[go: up one dir, main page]
More Web Proxy on the site http://driver.im/

CN113435313A - Pulse frequency domain feature extraction method based on DFT - Google Patents

Pulse frequency domain feature extraction method based on DFT Download PDF

Info

Publication number
CN113435313A
CN113435313A CN202110708487.9A CN202110708487A CN113435313A CN 113435313 A CN113435313 A CN 113435313A CN 202110708487 A CN202110708487 A CN 202110708487A CN 113435313 A CN113435313 A CN 113435313A
Authority
CN
China
Prior art keywords
dft
pulse
frequency domain
frequency
resolution
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202110708487.9A
Other languages
Chinese (zh)
Inventor
张蔚
杨大川
吕文超
易正红
黄辰
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
CETC 29 Research Institute
Southwest China Research Institute Electronic Equipment
Original Assignee
CETC 29 Research Institute
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by CETC 29 Research Institute filed Critical CETC 29 Research Institute
Priority to CN202110708487.9A priority Critical patent/CN113435313A/en
Publication of CN113435313A publication Critical patent/CN113435313A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
    • G06F2218/02—Preprocessing
    • G06F2218/04—Denoising
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/14—Fourier, Walsh or analogous domain transformations, e.g. Laplace, Hilbert, Karhunen-Loeve, transforms
    • G06F17/141—Discrete Fourier transforms
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
    • G06F2218/08—Feature extraction

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Theoretical Computer Science (AREA)
  • Mathematical Optimization (AREA)
  • Mathematical Analysis (AREA)
  • Data Mining & Analysis (AREA)
  • Pure & Applied Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Computational Mathematics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Artificial Intelligence (AREA)
  • Discrete Mathematics (AREA)
  • Algebra (AREA)
  • Signal Processing (AREA)
  • Databases & Information Systems (AREA)
  • Software Systems (AREA)
  • Analogue/Digital Conversion (AREA)

Abstract

The invention provides a pulse frequency domain feature extraction method based on DFT, which comprises the following steps: step 1, performing DFT (discrete Fourier transform) conversion of coarse resolution on a given pulse to obtain a power spectrum, obtaining a coarse center frequency on the power spectrum, and defining a bandwidth; step 2, selecting the center frequency, resolution and frequency range required by DFT conversion based on the rough center frequency obtained in the step 1; and 3, for a given pulse, performing DFT operation by using the defined bandwidth and the central frequency, resolution and frequency range selected in the step 2 to obtain the required pulse frequency domain characteristic. The method utilizes the characteristics that DFT can not be subjected to time length change in the frequency domain, can improve the signal-to-noise ratio and the like, solves the problems that inaccurate measurement is easy to carry out when pulse frequency domain characteristics are extracted, and simultaneously avoids the problems that windowing effect and frequency domain alignment are easy to introduce when FFT conversion is carried out at different sampling rates, pulse widths and the like; and also can reduce the amount of computation and memory space.

Description

Pulse frequency domain feature extraction method based on DFT
Technical Field
The invention relates to the technical field of pulse feature extraction, in particular to a pulse frequency domain feature extraction method based on DFT (discrete Fourier transform).
Background
Many characteristics of pulse signals are reflected in a pulse modulation domain, so that the pulse signals have special significance for the identification of a radiation source, and are concerned about in the fields of signal detection, parameter estimation, target classification and the like.
The pulse contains rich non-linear characteristics, such as pulse width, rising and falling edges, overshoot, inflection point and the like. These characteristics are difficult to estimate accurately, and may vary with different signal-to-noise ratios, amplitudes, widths, etc. of the pulses, and at the same time, different digital sampling lengths of the obtained signals are also caused by different signal detection and sampling rates in practical applications. In general, when parameter estimation and target classification are performed, the characteristics of the desired signal are as invariant as possible to the dynamic factors, so that various processing for transforming the signal from the time domain to the frequency domain is a good choice.
Mapping a time domain signal to a frequency domain based on Fast Fourier Transform (FFT), and controlling the resolution of the frequency domain by adjusting the number of FFT points in ways of increasing interpolation, zero padding and the like, but the methods bring signal distortion or window effect and are difficult to be applied to occasions with higher requirements on signal characteristic precision; for pulse feature extraction, the influence of links such as signal feature time domain and frequency domain alignment, normalization and the like is reduced.
Disclosure of Invention
The invention aims to provide a pulse frequency domain feature extraction method based on DFT, so as to solve the technical problems.
The invention provides a pulse frequency domain feature extraction method based on DFT, which comprises the following steps:
step 1, performing DFT (discrete Fourier transform) conversion of coarse resolution on a given pulse to obtain a power spectrum, obtaining a coarse center frequency on the power spectrum, and defining a bandwidth;
step 2, selecting the center frequency, resolution and frequency range required by DFT conversion based on the rough center frequency obtained in the step 1;
and 3, for a given pulse, performing DFT operation by using the defined bandwidth and the central frequency, resolution and frequency range selected in the step 2 to obtain the required pulse frequency domain characteristic.
Further, the method for performing full-frequency-domain DFT with coarse resolution on the given pulse in step 1 to obtain the power spectrum includes: performing DFT of coarse resolution on a given pulse X to convert the pulse X into a whole frequency domain aliasing period corresponding to a sampling rate, so as to obtain a frequency domain characteristic X (omega) of the pulse X; the power spectrum of the frequency domain feature X (ω) is then calculated.
Further, the method for obtaining the coarse center frequency on the power spectrum in step 1 comprises: the coarse center frequency is obtained by calculating the centroid of the power spectrum.
Further, the bandwidth in step 1 is directly defined as the 3dB bandwidth.
Further, in step 3, normalization needs to be performed on a given pulse before performing DFT operation.
Further, the method for normalizing the given pulse includes peak normalization or time-domain envelope normalization.
In summary, due to the adoption of the technical scheme, the invention has the beneficial effects that:
1. the method utilizes the characteristics that DFT can not be subjected to time length change in the frequency domain, can improve the signal-to-noise ratio and the like, solves the problems that inaccurate measurement is easy to carry out when pulse frequency domain characteristics are extracted, and simultaneously avoids the problems that windowing effect and frequency domain alignment are easy to introduce when FFT conversion is carried out at different sampling rates, pulse widths and the like.
2. The invention firstly utilizes the DFT of the coarse resolution to estimate the parameters of the pulse, thereby greatly saving the operation amount, being capable of adjusting the resolution and being more flexible in practical application.
3. The second DFT conversion is more targeted, only the frequency in the range with the pulse signal is operated, the storage space overhead and the operation resource overhead are reduced, the problem of time-frequency domain normalization brought by various factors in practical application can be solved, and the accuracy of image pulse frequency domain characteristics such as noise outside the pulse is reduced.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present invention, and therefore should not be considered as limiting the scope, and for those skilled in the art, other related drawings can be obtained according to the drawings without inventive efforts.
Fig. 1 is a flowchart of a DFT-based pulse frequency domain feature extraction method according to an embodiment of the present invention.
FIG. 2 is a flowchart of step 1 of an embodiment of the present invention.
FIG. 3 is a flowchart of step 3 of the present invention
Fig. 4 is a comparison graph of the time domain characteristics of two simulation signals in the FFT.
FIG. 5 is a graph of DFT signature comparison of two simulated signals at a specified frequency range and resolution.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. The components of embodiments of the present invention generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations.
Thus, the following detailed description of the embodiments of the present invention, presented in the figures, is not intended to limit the scope of the invention, as claimed, but is merely representative of selected embodiments of the invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Examples
As shown in fig. 1, this embodiment proposes a pulse frequency domain feature extraction method based on DFT, which includes the following steps:
step 1, performing DFT conversion of coarse resolution on a given pulse to obtain a power spectrum, obtaining a coarse center frequency on the power spectrum, and defining a bandwidth.
As shown in fig. 2, step 1 includes:
(1) and performing full-frequency-domain DFT conversion of coarse resolution on a given pulse to obtain a power spectrum. Specifically, for a given pulse X, performing DFT conversion of coarse resolution to the entire frequency domain aliasing period corresponding to the sampling rate, thereby obtaining a frequency domain feature X (ω) of the pulse X; the power spectrum of the frequency domain feature X (ω) is then calculated.
(2) And obtaining the rough center frequency on the power spectrum, wherein the rough center frequency can be obtained by calculating the mass center of the power spectrum.
(3) The bandwidth is defined, which is typically defined as a 3dB bandwidth.
Step 2, selecting the center frequency, resolution and frequency range required by DFT conversion based on the rough center frequency obtained in the step 1; the method specifically comprises the following steps: firstly, determining the center frequency f and resolution delta needed by DFT conversionf(ii) a And determining the frequency range BW required by the DFT transform, thereby determining discrete points of the output of the DFT transform, the discrete points being expressed as f + -n × δf(n*δf〈＝BW/2)。
And 3, for a given pulse, performing DFT operation by using the defined bandwidth and the central frequency, resolution and frequency range selected in the step 2 to obtain the required pulse frequency domain characteristic.
As shown in fig. 3, step 3 includes:
(1) for normalization of a given pulse, peak normalization or time-domain envelope normalization may be generally employed.
(2) And (3) performing DFT operation by using the defined bandwidth and the central frequency, resolution and frequency range selected in the step (2) to obtain the required pulse frequency domain characteristics.
In order to prove the effectiveness of the pulse frequency domain feature extraction method based on the DFT, the pulse frequency domain feature extraction method based on the DFT is subjected to a simulation test and is processed according to the processing flows of step 1, step 2 and step 3 shown in fig. 1, fig. 2 and fig. 3. Firstly, performing DFT conversion of coarse resolution on a given pulse to obtain a power spectrum, obtaining a coarse center frequency on the power spectrum, and defining a bandwidth; then, based on the rough center frequency obtained in the step 1, selecting the center frequency, the resolution and the frequency range required by DFT conversion; and finally, for a given pulse, performing DFT operation by using the defined bandwidth and the central frequency, resolution and frequency range selected in the step 2 to obtain the required pulse frequency domain characteristic.
The two simulated pulse signals are chirp pulse signals with the frequency of 100MHz and 125MHz respectively, the pulse width of 1us and 2us respectively, the bandwidth of 1MHz, and the signal sampling frequency of 500 MHz. The digital sample lengths obtained for different signals are different at this time.
FIG. 4 is a comparison of the features of two simulated signals in the FFT time domain, which is difficult to compare and distinguish; in fig. 4, the abscissa is frequency, corresponding to the full frequency domain, and the ordinate is power.
FIG. 5 is a comparison of DFT characteristics of two simulated signals at a given frequency range and resolution, with the pulses of two different parameters having a clear distinction; in fig. 5, the abscissa is the designated frequency range after DFT conversion, the actual center frequencies of different pulses are different, and the ordinate is power.
Therefore, the pulse frequency domain feature extraction method based on DFT can realize high-precision pulse feature characterization under the condition of noise and background interference, and proves the effectiveness of the invention. And the invention has the following beneficial effects:
1. the method utilizes the characteristics that DFT can not be subjected to time length change in the frequency domain, can improve the signal-to-noise ratio and the like, solves the problems that inaccurate measurement is easy to carry out when pulse frequency domain characteristics are extracted, and simultaneously avoids the problems that windowing effect and frequency domain alignment are easy to introduce when FFT conversion is carried out at different sampling rates, pulse widths and the like.
2. The invention firstly utilizes the DFT of the coarse resolution to estimate the parameters of the pulse, thereby greatly saving the operation amount, being capable of adjusting the resolution and being more flexible in practical application.
3. The second DFT conversion is more targeted, only the frequency in the range with the pulse signal is operated, the storage space overhead and the operation resource overhead are reduced, the problem of time-frequency domain normalization brought by various factors in practical application can be solved, and the accuracy of image pulse frequency domain characteristics such as noise outside the pulse is reduced.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims (6)

1. A pulse frequency domain feature extraction method based on DFT is characterized by comprising the following steps:
step 1, performing DFT (discrete Fourier transform) conversion of coarse resolution on a given pulse to obtain a power spectrum, obtaining a coarse center frequency on the power spectrum, and defining a bandwidth;
step 2, selecting the center frequency, resolution and frequency range required by DFT conversion based on the rough center frequency obtained in the step 1;
and 3, for a given pulse, performing DFT operation by using the defined bandwidth and the central frequency, resolution and frequency range selected in the step 2 to obtain the required pulse frequency domain characteristic.
2. The DFT-based pulse frequency domain feature extraction method as claimed in claim 1, wherein the step 1 of performing full-frequency domain DFT with coarse resolution on the given pulse to obtain the power spectrum comprises: performing DFT of coarse resolution on a given pulse X to convert the pulse X into a whole frequency domain aliasing period corresponding to a sampling rate, so as to obtain a frequency domain characteristic X (omega) of the pulse X; the power spectrum of the frequency domain feature X (ω) is then calculated.
3. The DFT-based pulse frequency domain feature extraction method as claimed in claim 1, wherein the method for obtaining the coarse center frequency on the power spectrum in step 1 comprises: the coarse center frequency is obtained by calculating the centroid of the power spectrum.
4. The DFT-based pulse frequency domain feature extraction method as recited in claim 1, wherein the bandwidth in step 1 is directly defined as a 3dB bandwidth.
5. The DFT-based pulse frequency domain feature extraction method as claimed in claim 1, wherein in step 3, normalization is performed on a given pulse before DFT operation is performed.
6. The DFT-based pulse frequency-domain feature extraction method of claim 5, wherein the method of normalizing a given pulse comprises peak normalization or time-domain envelope normalization.
CN202110708487.9A 2021-06-23 2021-06-23 Pulse frequency domain feature extraction method based on DFT Pending CN113435313A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110708487.9A CN113435313A (en) 2021-06-23 2021-06-23 Pulse frequency domain feature extraction method based on DFT

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110708487.9A CN113435313A (en) 2021-06-23 2021-06-23 Pulse frequency domain feature extraction method based on DFT

Publications (1)

Publication Number Publication Date
CN113435313A true CN113435313A (en) 2021-09-24

Family

ID=77754276

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110708487.9A Pending CN113435313A (en) 2021-06-23 2021-06-23 Pulse frequency domain feature extraction method based on DFT

Country Status (1)

Country Link
CN (1) CN113435313A (en)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP1626549A2 (en) * 2004-08-11 2006-02-15 Rohde & Schwarz GmbH & Co. KG Method and apparatus for estimating carrier frequency using a block-wise coarse estimation
CN103777199A (en) * 2014-02-24 2014-05-07 中国科学院电子学研究所 Distance measuring method of frequency modulated continuous wave radar system
CN107710323A (en) * 2016-01-22 2018-02-16 弗劳恩霍夫应用研究促进协会 Resampled using spectrum domain to encode or decode the device and method of audio multichannel signal
CN108509377A (en) * 2018-03-19 2018-09-07 东南大学 A kind of pulse signal arrival time based on along feature extraction and pulsewidth method of estimation

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP1626549A2 (en) * 2004-08-11 2006-02-15 Rohde & Schwarz GmbH & Co. KG Method and apparatus for estimating carrier frequency using a block-wise coarse estimation
CN103777199A (en) * 2014-02-24 2014-05-07 中国科学院电子学研究所 Distance measuring method of frequency modulated continuous wave radar system
CN107710323A (en) * 2016-01-22 2018-02-16 弗劳恩霍夫应用研究促进协会 Resampled using spectrum domain to encode or decode the device and method of audio multichannel signal
CN108509377A (en) * 2018-03-19 2018-09-07 东南大学 A kind of pulse signal arrival time based on along feature extraction and pulsewidth method of estimation

Similar Documents

Publication Publication Date Title
Aiello et al. A chirp-z transform-based synchronizer for power system measurements
US6230106B1 (en) Method of characterizing a device under test
CN109343020B (en) S-transform time-frequency analysis method based on improved window function
CN109471095B (en) FMCW radar distance estimation method based on fast iterative interpolation
CN107085140B (en) Nonequilibrium system frequency estimating methods based on improved SmartDFT algorithm
CN107315714B (en) Deconvolution power spectrum estimation method
CN112051583A (en) Beat frequency signal nonlinear correction method in FMCW distance measurement system
CN113156206B (en) Time-frequency combined noise-containing signal parameter estimation new algorithm
TWI405979B (en) Probability density function separating apparatus, probability density function separating method, noise separating apparatus, noise separating method, testing apparatus, testing method, calculating apparatus, calculating method, program, and recording m
Belega et al. Accurate amplitude estimation of harmonic components of incoherently sampled signals in the frequency domain
RU2435168C1 (en) Method for harmonic analysis of periodic multifrequency signal
Sato et al. Revisit to accurate ADC testing with incoherent sampling using proper sinusoidal signal and sampling frequencies
CN105282086A (en) Double-tone quadrature measuring method for frequency conversion system group delay
CN113435313A (en) Pulse frequency domain feature extraction method based on DFT
CN113406386B (en) Signal frequency accurate estimation method based on digital down-conversion
CN113740834B (en) Difference frequency estimation method of FMCW radar
Attivissimo et al. A study on nonlinear averagings to perform the characterization of power spectral density estimation algorithms
CN112485521B (en) Frequency estimation method based on symmetric discrete Fourier transform
Sedlacek et al. Active power measurements-an overview and comparison of DSP algorithms by noncoherent sampling
CN114035149A (en) Method for improving sensitivity of interferometer direction finding system
CN116996137B (en) Low signal-to-noise ratio broadband linear frequency modulation signal detection method based on weighted superposition
Wolf et al. Amplitude and frequency estimator for aperiodic multi-frequency noisy vibration signals of a tram gearbox
US6873923B1 (en) Systems and methods for performing analysis of a multi-tone signal
KR101935991B1 (en) Extreme fine frequency estimation apparatus and method of single receiver
CN113702969B (en) Micro Doppler signal parameter estimation method based on self-adaptive STFT method

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
RJ01 Rejection of invention patent application after publication
RJ01 Rejection of invention patent application after publication

Application publication date: 20210924