CN112929053B - Frequency hopping signal feature extraction and parameter estimation method - Google Patents
Frequency hopping signal feature extraction and parameter estimation method Download PDFInfo
- Publication number
- CN112929053B CN112929053B CN202110258849.9A CN202110258849A CN112929053B CN 112929053 B CN112929053 B CN 112929053B CN 202110258849 A CN202110258849 A CN 202110258849A CN 112929053 B CN112929053 B CN 112929053B
- Authority
- CN
- China
- Prior art keywords
- frequency hopping
- frequency
- time
- spwvd
- signal
- 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.)
- Active
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B1/00—Details of transmission systems, not covered by a single one of groups H04B3/00 - H04B13/00; Details of transmission systems not characterised by the medium used for transmission
- H04B1/69—Spread spectrum techniques
- H04B1/713—Spread spectrum techniques using frequency hopping
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B1/00—Details of transmission systems, not covered by a single one of groups H04B3/00 - H04B13/00; Details of transmission systems not characterised by the medium used for transmission
- H04B1/69—Spread spectrum techniques
- H04B1/713—Spread spectrum techniques using frequency hopping
- H04B1/715—Interference-related aspects
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Complex Calculations (AREA)
Abstract
The invention discloses a frequency hopping signal feature extraction and parameter estimation method, which overcomes the problems that the prior art has mutual restriction of time and frequency resolution, cross terms and TT transformation are sensitive to noise, and comprises the following steps: 1. sampling a frequency hopping signal; 2. performing smooth pseudo Wigner time frequency transformation; 3. TT transformation based on SPWVD transformation is carried out; 4. drawing an SPWVD-TT transformation graph; 5. extracting contour lines of the outermost circle of the eye-shaped structure in the obtained SPWVD-TT conversion graph; 6. solving an envelope curve, and removing a direct current component; 7. estimating a frequency hopping period: performing Fourier transform on the upper envelope curve without the direct-current component, wherein the frequency corresponding to the maximum amplitude value in the obtained amplitude-frequency diagram is the estimated value of the frequency hopping rate, and the reciprocal of the frequency hopping rate is the estimated value of the frequency hopping period; 8. estimating the frequency hopping time: passing the remaining peak time points t1,t2,...,tk‑1Estimating the frequency hopping time; the error is reduced by an accumulation average method, and the estimated value of the initial frequency hopping time is
Description
Technical Field
The present invention relates to the field of signal processing technologies, and in particular, to a method for extracting characteristics and estimating parameters of a frequency hopping signal.
Background
The frequency hopping communication has good anti-interference, low interception probability and flexible networking capability, so that the frequency hopping communication is greatly developed in the military field, and various radio stations adopting the frequency hopping technology are widely applied in the military field, so that the interception resistance and the anti-interference capability of military equipment are greatly improved. In communication countermeasure, efficient extraction of parameters of frequency hopping communication signals is a prerequisite for implementing interference. Therefore, parameter estimation of frequency hopping signals with unknown parameters has become a key point of research on communication countermeasure and has important significance on modern military and national defense safety.
The frequency hopping signal refers to a carrier wave of a transmission signal hopping according to a frequency hopping sequence, has a time-varying and pseudo-random carrier frequency, and is a typical multi-component non-stationary signal. The conventional fourier transform method cannot analyze the frequency distribution of the frequency hopping signal at a specific time and the change of the frequency with time. Therefore, a time-frequency analysis method is usually adopted to analyze and estimate parameters of the frequency hopping signal. The main time-frequency analysis methods are classified into linear time-frequency analysis methods and nonlinear time-frequency analysis methods. The linear time-frequency analysis method mainly comprises the following steps: short-time fourier transform, Gabor transform, wavelet transform, S transform, etc. But the time resolution and the frequency resolution are restricted by the Heisenberg inaccurate measurement principle. Compared with a linear time frequency analysis method, the nonlinear time frequency analysis method has higher time frequency resolution, but is interfered by cross terms, and influences are caused on the analysis of useful signals.
TT transformation is a time-time analysis method based on S transformation. Compared with a time-frequency analysis method, the method has the following characteristics: the method has good frequency aggregation capability, can suppress low-frequency components of the original signal and accurately capture the initial moment of signal mutation. The TT transform can therefore be used as a tool to detect signal frequency changes and frequency hopping signal parameter estimation. However, the most serious problem of the transformation is that the transformation is sensitive to noise, and the parameters of the frequency hopping signal cannot be estimated when a large amount of noise exists.
Disclosure of Invention
The invention aims to solve the technical problems that the traditional time-frequency analysis method has the mutual restriction of time and frequency resolution, cross terms exist and TT transformation is sensitive to noise, and provides a frequency hopping signal feature extraction and parameter estimation method.
In order to solve the technical problems, the invention is realized by adopting the following technical scheme: the method for extracting the characteristics of the frequency hopping signal and estimating the parameters comprises the following steps:
1) sampling a frequency hopping signal;
2) performing smooth pseudo Wigner time frequency transformation;
3) performing TT conversion based on SPWVD:
the inverse Fourier transform about the frequency f is solved for the smooth pseudo Wigner time frequency transform result, and the expression is as follows
In the formula: w (t, tau) is the inverse Fourier transform result of the smooth pseudo Wigner time frequency transform; t is a time variable; f is a frequency variable; τ is a time delay;
4) drawing an SPWVD-TT transformation graph;
5) extracting contour lines of the outermost circle of an eye-shaped structure in the obtained SPWVD-TT transformation graph:
the eye-shaped structure in the SPWVD-TT conversion graph is formed by surrounding a plurality of contour lines, and the contour lines are gradually increased from the outermost layer to the innermost layer; the contour lines of the outermost layer have the smallest value, the contour lines of the innermost layer have the largest value, meanwhile, the contour lines gradually increase from the outermost layer to the innermost layer, the contour lines of the outermost layer can reflect the outline of an eye-shaped structure, namely the hopping of the frequency hopping signal frequency, and therefore the contour lines of the outermost circle of the eye-shaped structure in the SPWVD-TT conversion graph are extracted and used for estimating the parameters of the frequency hopping signal;
the method for extracting contour lines of the outermost circle of the eye-shaped structure in the SPWVD-TT transformation graph comprises the following steps: searching a contour line with the abscissa range of the minimum median value of contour lines from 1 to N through an output matrix of a contour function in MATLAB, namely, the contour line of the outermost circle of the eye-shaped structure in the SPWVD-TT transformation graph;
6) solving an envelope curve, and removing a direct current component;
7) estimating a frequency hopping period;
8) the hopping instants are estimated.
The frequency hopping signal sampling in the technical scheme is that the time domain sampling is carried out on the acquired frequency hopping signal, and the sampling frequency is fsObtaining a discrete sampling signal x (N), wherein N is 1, 2. And N is the number of sampling points.
The technical scheme of performing smooth pseudo Wigner time frequency transformation refers to the following steps: performing smooth pseudo Wigner time frequency transformation on the sampling signal x (n), wherein the expression of the smooth pseudo Wigner time frequency transformation is
In the formula:
a smooth pseudo-Wigner time-frequency transform for signal x (t); t is a time variable; f is a frequency variable; τ is a time delay; u is an integral variable; x is the number of*Is the conjugate of x; h (τ) and G (u) are odd-length window functions, and satisfy h (0) ═ G (0) ═ 1, where h (0) denotes the value of the window function h (τ) at 0, and G (0) denotes the value of the fourier transform of the window function G (u) at 0.
The drawing of the SPWVD-TT transformation graph in the technical scheme refers to: the SPWVD-TT conversion graph is a contour graph obtained by drawing a contour function in MATLAB through an SPWVD-TT conversion result, when a frequency hopping signal generates frequency hopping once and lasts for a complete frequency hopping period, an equal-size eye-shaped structure can be generated, the size of the structure is irrelevant to the size of the frequency hopping frequency and only relevant to the duration of a single frequency point, and the complete duration of each frequency point of the frequency hopping signal is one frequency hopping period, so that the structure of the frequency hopping signal has periodicity;
after a plurality of frequency hopping periods, a plurality of eye-shaped structures are presented on a symmetrical axis of the SPWVD-TT conversion graph in an end-to-end connection state, the number of the eye-shaped structures is equal to the number of the frequency hopping periods, and the frequency hopping signals can be distinguished from other communication signals by utilizing the periodicity of the SPWVD-TT conversion graph of the frequency hopping signals and the equal size of the eye-shaped structures, so that the time-time domain characteristic can be used as an identification characteristic of the frequency hopping signals.
In the technical scheme, the step of solving the upper envelope curve and the step of removing the direct current component refers to the following steps: calculating an upper envelope curve of the contour line, wherein the upper envelope curve is recorded as m (t), and the method comprises the following steps: searching a maximum ordinate corresponding to an abscissa k (k is 1, 2.., N) on the contour line, and taking the maximum ordinate as a ordinate corresponding to the abscissa k in the upper envelope curve; traversing all the abscissa values to obtain an upper envelope curve;
removing the direct current component of m (t), and recording the upper envelope curve of the removed direct current component as m1(t); removing a direct current component, and expressing the following expression:
in the formula, m1(t) is the upper envelope curve after the direct current component is removed; m (t) is an upper envelope curve;
is the mean value of m (t).
The estimation of the frequency hopping period in the technical scheme refers to: for the upper envelope curve m after the direct current component is removed1(t) performing a fourier transform, the fourier transform expression being:
wherein: f (f) is m1(t) Fourier transform; f is the frequency;
the frequency corresponding to the maximum amplitude value in the obtained amplitude-frequency diagram is the estimated value of the frequency hopping rate
Obtaining the estimated value of the frequency hopping period by taking the reciprocal of the obtained frequency hopping rate
I.e. the number of samples contained in a hop period.
The estimation of the frequency hopping time in the technical scheme refers to: searching the upper envelope curve m with the DC component removed1(t) all peak time points, respectively t0,t1,...tk(ii) a In most cases, the first hop and the last hop of a received signal are incomplete, which affects the estimation of frequency hopping time, so that the first peak time point t and the last peak time point t are omitted0,tkPassing the remaining peak time point t1,t2,...,tk-1Estimating the frequency hopping time;
the error is reduced by an accumulation average method, and the estimated value of the initial frequency hopping time is
The nth frequency hopping time estimated value is
Wherein
Is an estimate of the frequency hopping period.
Compared with the prior art, the invention has the beneficial effects that:
1. according to the frequency hopping signal feature extraction and parameter estimation method, inverse Fourier transform about frequency is solved for smooth pseudo Wigner distribution, frequency hopping signal features are reflected on a time-time domain, when frequency hopping occurs once and a complete frequency hopping period continues, an equal-size eye-shaped structure can be generated, the size of the structure is irrelevant to the frequency hopping frequency and only relevant to the duration time of a single frequency point, and the complete duration time of each frequency point of the frequency hopping signal is one frequency hopping period, so that the structure of the frequency hopping signal has periodicity, the time-time domain features can be used as identification features of the frequency hopping signal, and the frequency hopping signal can be distinguished from other communication signals through the features;
2. the frequency hopping signal feature extraction and parameter estimation method can estimate the frequency hopping period, frequency hopping rate and frequency hopping time of the frequency hopping signal through the periodicity of the eye-shaped structure, is different from a common frequency hopping signal parameter estimation method based on time-frequency analysis, belongs to the field of time-time domain signal analysis, provides a feasible new method for the parameter estimation problem of the frequency hopping signal, and effectively avoids the problems of mutual restriction of time and frequency resolution and cross terms existing in the traditional time-frequency analysis, which are difficult to solve;
3. the SPWVD-TT conversion has better noise immunity than TT conversion, frequency jump of a frequency hopping signal can be detected even under the condition of low signal to noise ratio, parameters of the frequency hopping signal are estimated, and the problems that the TT conversion is sensitive to noise, frequency change cannot be detected under a large amount of noise environments, and the parameters are estimated are effectively solved.
Drawings
The invention is further described with reference to the accompanying drawings in which:
fig. 1 is a flow chart of a method for extracting characteristics of a frequency hopping signal and estimating parameters according to the present invention;
FIG. 2 is a time domain waveform diagram of a frequency hopping signal of 0dB SNR according to the present invention;
FIG. 3 is a diagram of SPWVD-TT transformation of a frequency hopping signal under a noise-free condition according to the present invention;
FIG. 4 is a diagram of SPWVD-TT transformation of a frequency hopping signal with a signal-to-noise ratio of 0dB according to the present invention;
FIG. 5 is a TT transformation diagram of a frequency hopping signal under a noise-free condition provided by the present invention;
FIG. 6 is a TT transformation plot of a frequency hopping signal with a signal-to-noise ratio of 0dB according to the present invention;
FIG. 7 is a time domain waveform diagram of a frequency hopping signal TT transforming diagonal signal when the signal-to-noise ratio provided by the present invention is 0 dB;
FIG. 8 is a contour diagram of the lowest numerical value in the lower half of the symmetry axis of the SPWVD-TT conversion diagram of the frequency hopping signal provided by the present invention;
FIG. 9 is a graph of the upper envelope of the present invention with the DC component removed;
fig. 10 is an amplitude-frequency diagram obtained by performing fourier transform on the upper envelope curve from which the dc component is removed according to the present invention;
Detailed Description
The invention is described in detail below with reference to the attached drawing figures:
referring to fig. 1, when performing parameter estimation on a frequency hopping signal, the method for extracting characteristics of the frequency hopping signal and estimating parameters is provided in the present invention, aiming at the problems that the conventional time-frequency analysis method has mutual restriction of time and frequency resolution and has cross terms, and TT transformation is sensitive to noise and cannot perform parameter estimation on the frequency hopping signal under the condition of low signal-to-noise ratio, and the method includes the following steps:
1. frequency hopping signal sampling
Carrying out time domain sampling on a frequency hopping signal to be subjected to parameter estimation, wherein the sampling frequency is fsObtaining a discrete sampling signal sequence x (n) of the frequency hopping signal; n is 1, 2.. No. N; n is the number of sampling points;
2. making smooth pseudo Wigner time-frequency transformation
The sampling signal x (n) is subjected to smooth pseudo Wigner time frequency transformation (SPWVD), which is a quadratic time frequency analysis method, not only has good time frequency aggregation, but also has certain capability of inhibiting cross terms and noise interference in time frequency distribution, and the expression is
In the formula:
a smooth pseudo-Wigner time-frequency transform for signal x (t); t is a time variable; f is a frequency variable; τ is a time delay; u is an integral variable; x is the number of*Is the conjugate of x; h (τ) and G (u) are odd-length window functions, satisfying h (0) ═ G (0) ═ 1, where h (0) denotes the value of the window function h (τ) at 0, and G (0) denotes the value of the fourier transform of the window function G (u) at 0;
3. TT transformation based on SPWVD (SPWVD-TT)
The inverse Fourier transform about the frequency f is solved for the smooth pseudo Wigner time frequency transform result, and the expression is as follows
In the formula: w (t, tau) is the inverse Fourier transform result of the smooth pseudo Wigner time frequency transform; t is a time variable; f is a frequency variable; τ is a time delay;
4. drawing SPWVD-TT conversion graph
Drawing an SPWVD-TT transformation graph of the sampling signal x (n) to obtain time-time domain characteristics of the sampling signal x (n);
the SPWVD-TT conversion graph is a contour graph obtained by drawing a contour function in MATLAB through an SPWVD-TT conversion result, when a frequency hopping signal generates frequency hopping once and lasts for a complete frequency hopping period, an equal-size eye-shaped structure can be generated, the size of the structure is irrelevant to the size of the frequency hopping frequency and only relevant to the duration of a single frequency point, and the complete duration of each frequency point of the frequency hopping signal is one frequency hopping period, so that the structure of the frequency hopping signal has periodicity; after a plurality of frequency hopping periods, a plurality of eye-shaped structures are presented on a symmetrical axis of the SPWVD-TT conversion graph in an end-to-end connection state, the number of the eye-shaped structures is equal to the number of the frequency hopping periods, and the frequency hopping signals can be distinguished from other communication signals by utilizing the periodicity of the SPWVD-TT conversion graph of the frequency hopping signals and the equal size of the eye-shaped structures, so that the time-time domain characteristic can be used as an identification characteristic of the frequency hopping signals;
5. contour lines of the outermost circle of the eye-shaped structure in the SPWVD-TT transformation graph obtained by extraction
The eye-shaped structure in the SPWVD-TT conversion graph is formed by surrounding a plurality of contour lines, and the contour lines are gradually increased from the outermost layer to the innermost layer; the contour lines of the outermost layer have the smallest value, the contour lines of the innermost layer have the largest value, meanwhile, the contour lines gradually increase from the outermost layer to the innermost layer, the contour lines of the outermost layer can reflect the outline of an eye-shaped structure, namely the hopping of the frequency hopping signal frequency, and therefore the contour lines of the outermost circle of the eye-shaped structure in the SPWVD-TT conversion graph are extracted and used for estimating the parameters of the frequency hopping signal;
the method for extracting contour lines of the outermost circle of the eye-shaped structure in the SPWVD-TT transformation graph comprises the following steps: searching a contour line with the abscissa range of the minimum median value of contour lines from 1 to N through an output matrix of a contour function in MATLAB, namely, the contour line of the outermost circle of the eye-shaped structure in the SPWVD-TT transformation graph;
6. calculating an envelope curve to remove DC component
Calculating an upper envelope curve of the contour line, and recording the upper envelope curve as m (t); the method for solving the upper envelope curve of the contour line comprises the following steps: searching a maximum ordinate corresponding to an abscissa k (k is 1, 2.., N) on the contour line, and taking the maximum ordinate as a ordinate corresponding to the abscissa k in the upper envelope curve; and traversing all the abscissa values to obtain an upper envelope curve.
Removing the direct current component of m (t), and recording the upper envelope curve of the removed direct current component as m1(t); removing a direct current component, and expressing the following expression:
in the formula, m1(t) is the upper envelope curve after the direct current component is removed; m (t) is an upper envelope curve;
is the mean value of m (t);
7. estimating a frequency hopping period
For the upper envelope curve m after the direct current component is removed1(t) performing a fourier transform, the fourier transform expression being:
wherein: f (f) is m1(t) Fourier transform; f is the frequency;
the frequency corresponding to the maximum amplitude value in the obtained amplitude-frequency diagram is the estimated value of the frequency hopping rate
Obtaining the estimated value of the frequency hopping period by taking the reciprocal of the obtained frequency hopping rate
The number of sampling points contained in one frequency hopping period is obtained;
8. estimating frequency hopping time
Searching the upper envelope curve m with the DC component removed1(t) all peak time points, respectively t0,t1,...tk(ii) a In most cases, the first hop and the last hop of a received signal are incomplete, which affects the estimation of frequency hopping time, so that the first peak time point t and the last peak time point t are omitted0,tkPassing the remaining peak time point t1,t2,...,tk-1Estimating the frequency hopping time;
the error is reduced by an accumulation average method, and the estimated value of the initial frequency hopping time is
The nth frequency hopping time estimated value is
Wherein
Is an estimate of the frequency hopping period.
In order to make those skilled in the art better understand the reconstruction method described in the present embodiment, the reconstruction method is described below with reference to a specific example.
Simulation conditions are as follows: the sampling point number of the frequency hopping signal is 2048, the sampling frequency is 2048kHz, the signal duration is 1ms, the carrier amplitude is 1, the hopping speed is 8000hop/s, the sampling point number is 256 in one frequency hopping period, the frequency hopping frequency set is {300,500,600,400,100,800,200,700,300} kHz, wherein the first hop and the last hop are not complete frequency hopping periods, the sampling point number is 150 and 106 respectively, and the signal-to-noise ratio is 0dB under the condition of Gaussian white noise;
1. frequency hopping signal sampling
Carrying out time domain sampling on the simulation signal, wherein the sampling frequency is 2048kHz, and the number of sampling points is 2048, so as to obtain a sampling sequence x (n); n 1, 2.... 2048; as shown in fig. 2, which is a time domain waveform diagram of a frequency hopping signal when the signal-to-noise ratio is 0dB, it can be seen that in a noise environment, parameter information of the frequency hopping signal cannot be obtained through the time domain waveform diagram;
2. making smooth pseudo Wigner time-frequency transformation
Performing smooth pseudo Wigner time frequency transformation (SPWVD) on the sampling signal x (n), wherein the expression of the smooth pseudo Wigner time frequency transformation is
In the formula:
is the smooth pseudo-Wigner time-frequency transformation of the signal x (t), t being the time variable, f being the frequency variable, τ being the time delay, u being the integral variable, x*For the conjugate of x, h (τ) and G (u) are odd-length window functions, satisfying h (0) ═ G (0) ═ 1, where h (0) denotes the value of the window function h (τ) at 0, and G (0) denotes the value of the fourier transform of the window function G (u) at 0.
3. TT transformation based on SPWVD (SPWVD-TT)
The inverse Fourier transform about the frequency f is solved for the smooth pseudo Wigner time frequency transform result, and the expression is as follows
W (t, tau) is the inverse Fourier transform result of smooth pseudo Wigner time frequency transformation; t is a time variable, f is a frequency variable, and τ is a time delay;
4. drawing SPWVD-TT conversion graph
Drawing an SPWVD-TT transformation graph of a sampling signal x (n) by utilizing a contour function in MATLAB through an SPWVD-TT transformation result to obtain a time-time domain characteristic of the sampling signal x (n), wherein as shown in FIG. 3, the time-time domain characteristic is an SPWVD-TT transformation graph of a frequency hopping signal under a noise-free condition, FIG. 4 is an SPWVD-TT transformation graph of the frequency hopping signal when the signal-to-noise ratio is 0dB, and the comparison between the graphs in FIGS. 3 and 4 is visible, so that a large amount of noise can not greatly influence the time-time domain characteristic, namely, the time-time domain signal characteristic based on the SPWVD-TT transformation has stronger anti-noise performance; in order to highlight the noise immunity of the SPWVD-TT transformation, the TT transformation result is compared with that, as shown in fig. 5, it is a TT transformation graph of a frequency hopping signal under a noise-free condition, fig. 6 is a TT transformation graph of a frequency hopping signal when the signal-to-noise ratio is 0dB, and fig. 7 is a time domain waveform graph of a signal on a diagonal line of the TT transformation.
As shown in fig. 3 and 4, when the frequency hopping signal goes through a complete hop, it appears on the SPWVD transformation diagram to generate an equally large "eye" structure, which is independent of the frequency hopping frequency and only dependent on the frequency hopping period; the eye-shaped structures are connected with each other, have obvious periodicity and can reflect the parameter information of the frequency hopping signal. The isochronism and periodicity of the "eye" structure, as well as noise robustness, allows frequency hopping signals to be identified by this time-time domain signature.
5. Contour lines of the outermost circle of the eye-shaped structure in the SPWVD-TT transformation graph obtained by extraction
The SPWVD-TT conversion graph of the frequency hopping signal is a graph which is symmetrical up and down, the contour line with the minimum value in the graph is positioned on the outermost layer, and the contour line with the minimum value in the contour line with the horizontal coordinate ranging from 1 to 2048 is searched through an output matrix of a contour function in MATLAB, namely the contour line of the outermost ring of the eye-shaped structure in the SPWVD-TT conversion graph; contour lines of the outermost circle below the symmetry axis of the SPWVD-TT conversion diagram of the frequency hopping signal are extracted, as shown in FIG. 8, the contour lines with the minimum numerical value in the graph below the symmetry axis of the SPWVD-TT conversion diagram of the frequency hopping signal are obvious, and the contour lines have obvious periodicity.
6. Calculating an envelope curve to remove DC component
Drawing an upper envelope curve m (t) of a contour line with the minimum numerical value in a graph below a symmetric axis of a frequency hopping signal SPWVD-TT conversion graph, wherein the method for solving the upper envelope curve of the contour line comprises the following steps: searching a maximum ordinate corresponding to an abscissa k (k is 1, 2.., 2048) on the contour line, and taking the maximum ordinate as an ordinate corresponding to the abscissa k in the upper envelope curve; and traversing all the abscissa values to obtain an upper envelope curve.
Calculating the average value of the upper envelope curve m (t), and removing the direct current component, wherein the expression is as follows:
in the formula: m is1(t) is the upper envelope curve after the direct current component is removed; m (t) is an upper envelope curve;
is the mean value of m (t).
As shown in fig. 9, in order to remove the upper envelope curve after the dc component, it can be seen that the curve also has obvious periodicity, which reflects the parameter information of the frequency hopping signal;
7. estimating a frequency hopping period
To m1(t) performing a fourier transform, the fourier transform expression being:
wherein F (f) is m1(t) Fourier transform; f is the frequency.
The frequency corresponding to the maximum amplitude value in the obtained amplitude-frequency diagram is the estimated value of the frequency hopping rate
Obtaining the estimated value of the frequency hopping period by taking the reciprocal of the obtained frequency hopping rate
I.e. the number of samples contained in a hop period.
As shown in fig. 10, the amplitude-frequency diagram is obtained by performing fourier transform on the upper envelope curve from which the dc component is removed; from this, the frequency hopping rate estimate is
The frequency hopping period estimate is
The number of sampling points in one frequency hopping period is 0.125ms 2048kHz 256, and the sampling points are completely consistent with the real parameter values;
8. estimating frequency hopping time
Searching the upper envelope curve m after removing the direct current component1(t) all peak sample points, 145,413,662,927,1168,1425,1685,1966 respectively; because the first hop and the last hop of the received signal are incomplete and influence the estimation of the frequency hopping moment, the first peak value sampling point 145 and the last peak value sampling point 1966 are omitted, and the frequency hopping moment is estimated through the rest peak value sampling points;
the error is reduced by an accumulation average method, and the calculation expression of the estimated value of the initial frequency hopping time is
The initial frequency hopping time estimation result is
Actual value of t0＝0.0732ms
The nth frequency hopping time estimated value calculation expression is
the frequency hopping time estimation sequence is as follows:
0.0736ms,0.1986ms,0.3236ms,0.4486ms,0.5736ms,0.6986ms,0.8236ms,0.9486ms
the actual sequence of the hopping time is:
0.0732ms,0.1982ms,0.3232ms,0.4482ms,0.5732ms,0.6982ms,0.8232ms,0.9482ms；
the difference between the frequency hopping time estimation sequence and the real sequence is only 0.0004ms, and the estimation precision is good.
The feasibility and the estimation accuracy of the time-time domain frequency hopping signal parameter estimation method based on the SPWVD-TT transformation are verified. The method performs feature extraction and parameter estimation on the frequency hopping signal from the time-time domain angle, provides a new approach for frequency hopping signal identification and parameter estimation, effectively avoids the problems that the time resolution and the frequency resolution are mutually restricted and cross interference exists on the time-frequency domain, and effectively solves the problem that TT transformation cannot realize parameter estimation of the signal under the condition of a large amount of noise.
Claims (7)
1. A frequency hopping signal feature extraction and parameter estimation method is characterized by comprising the following steps:
1) sampling a frequency hopping signal;
2) performing smooth pseudo Wigner time frequency transformation;
3) performing TT conversion based on SPWVD:
the inverse Fourier transform about the frequency f is solved for the smooth pseudo Wigner time frequency transform result, and the expression is as follows
In the formula: w (t, tau) is the inverse Fourier transform result of the smooth pseudo Wigner time frequency transform; t is a time variable; f is a frequency variable; τ is a time delay;
4) drawing an SPWVD-TT transformation graph;
5) extracting contour lines of the outermost circle of an eye-shaped structure in the obtained SPWVD-TT transformation graph:
the eye-shaped structure in the SPWVD-TT conversion graph is formed by surrounding a plurality of contour lines, and the contour lines are gradually increased from the outermost layer to the innermost layer; the contour lines of the outermost layer have the smallest value, the contour lines of the innermost layer have the largest value, meanwhile, the contour lines gradually increase from the outermost layer to the innermost layer, the contour lines of the outermost layer can reflect the outline of an eye-shaped structure, namely the hopping of the frequency hopping signal frequency, and therefore the contour lines of the outermost circle of the eye-shaped structure in the SPWVD-TT conversion graph are extracted and used for estimating the parameters of the frequency hopping signal;
the method for extracting contour lines of the outermost circle of the eye-shaped structure in the SPWVD-TT transformation graph comprises the following steps: searching a contour line with the abscissa range of the minimum median value of contour lines from 1 to N through an output matrix of a contour function in MATLAB, namely, the contour line of the outermost circle of the eye-shaped structure in the SPWVD-TT transformation graph;
6) solving an envelope curve, and removing a direct current component;
7) estimating a frequency hopping period;
8) the hopping instants are estimated.
2. The method of claim 1, wherein the frequency hopping signal sampling comprises:
carrying out time domain sampling on the acquired frequency hopping signal, wherein the sampling frequency is fsObtaining a discrete sampling signal x (N), wherein N is 1, 2. And N is the number of sampling points.
3. The method for extracting characteristics and estimating parameters of a frequency hopping signal according to claim 1, wherein the performing of the smooth pseudo-Wigner time-frequency transformation is:
performing smooth pseudo Wigner time frequency transformation on the sampling signal x (n), wherein the expression of the smooth pseudo Wigner time frequency transformation is
In the formula:
a smooth pseudo-Wigner time-frequency transform for signal x (t); t is a time variable; f is a frequency variable; τ is a time delay; u is an integral variable; x is the number of*Is the conjugate of x; h (τ) and G (u) are odd-length window functions, and satisfy h (0) ═ G (0) ═ 1, where h (0) denotes the value of the window function h (τ) at 0, and G (0) denotes the value of the fourier transform of the window function G (u) at 0.
4. The method of claim 1, wherein the drawing of the SPWVD-TT transformation graph comprises:
the SPWVD-TT conversion graph is a contour graph obtained by drawing a contour function in MATLAB through an SPWVD-TT conversion result, when a frequency hopping signal generates frequency hopping once and lasts for a complete frequency hopping period, an equal-size eye-shaped structure can be generated, the size of the structure is irrelevant to the size of the frequency hopping frequency and only relevant to the duration of a single frequency point, and the complete duration of each frequency point of the frequency hopping signal is one frequency hopping period, so that the structure of the frequency hopping signal has periodicity;
after a plurality of frequency hopping periods, a plurality of eye-shaped structures are presented on a symmetrical axis of the SPWVD-TT conversion graph in an end-to-end connection state, the number of the eye-shaped structures is equal to the number of the frequency hopping periods, and the frequency hopping signals can be distinguished from other communication signals by utilizing the periodicity of the SPWVD-TT conversion graph of the frequency hopping signals and the equal size of the eye-shaped structures, so that the time-time domain characteristic can be used as an identification characteristic of the frequency hopping signals.
5. The method of claim 1, wherein the step of finding the envelope curve and removing the dc component comprises:
calculating an upper envelope curve of the contour line, wherein the upper envelope curve is recorded as m (t), and the method comprises the following steps:
searching a maximum ordinate corresponding to an abscissa k (k is 1, 2.., N) on the contour line, and taking the maximum ordinate as a ordinate corresponding to the abscissa k in the upper envelope curve; traversing all the abscissa values to obtain an upper envelope curve;
removing the direct current component of m (t), and recording the upper envelope curve of the removed direct current component as m1(t); removing a direct current component, and expressing the following expression:
in the formula, m1(t) is the upper envelope curve after the direct current component is removed; m (t) is an upper envelope curve;
is the mean value of m (t).
6. The method of claim 1, wherein the estimating the hop period comprises:
for the upper envelope curve m after the direct current component is removed1(t) performing a fourier transform, the fourier transform expression being:
wherein: f (f) is m1(t) Fourier transform; f is the frequency;
the frequency corresponding to the maximum amplitude value in the obtained amplitude-frequency diagram is the estimated value of the frequency hopping rate
Obtaining the estimated value of the frequency hopping period by taking the reciprocal of the obtained frequency hopping rate
I.e. the number of samples contained in a hop period.
7. The method of claim 1, wherein the estimating the hopping time comprises:
searching the upper envelope curve m with the DC component removed1(t) all peak time points, respectively t0,t1,...tk(ii) a In most cases, the first hop and the last hop of a received signal are incomplete, which affects the estimation of frequency hopping time, so that the first peak time point t and the last peak time point t are omitted0,tkPassing the remaining peak time point t1,t2,...,tk-1Estimating the frequency hopping time;
the error is reduced by an accumulation average method, and the estimated value of the initial frequency hopping time is
The nth frequency hopping time estimated value is
Wherein
Is an estimate of the frequency hopping period.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110258849.9A CN112929053B (en) | 2021-03-10 | 2021-03-10 | Frequency hopping signal feature extraction and parameter estimation method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110258849.9A CN112929053B (en) | 2021-03-10 | 2021-03-10 | Frequency hopping signal feature extraction and parameter estimation method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112929053A CN112929053A (en) | 2021-06-08 |
| CN112929053B true CN112929053B (en) | 2022-02-22 |
Family
ID=76172283
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110258849.9A Active CN112929053B (en) | 2021-03-10 | 2021-03-10 | Frequency hopping signal feature extraction and parameter estimation method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112929053B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102305940A (en) * | 2011-05-24 | 2012-01-04 | 中国石油集团川庆钻探工程有限公司 | Method for extracting fluid factor |
| CN105991492A (en) * | 2015-02-10 | 2016-10-05 | 北京空间飞行器总体设计部 | Frequency hopping (FH) signal identification method |
| CN107276629A (en) * | 2017-06-20 | 2017-10-20 | 中国人民解放军军械工程学院 | Hop cycle method of estimation based on time frequency analysis center of gravity of frequency |
| CN110113075A (en) * | 2018-10-29 | 2019-08-09 | 西安电子科技大学 | Hybrid network platform frequency parameter blind estimating method based on STFT-SPWVD |
| CN111934711A (en) * | 2020-06-30 | 2020-11-13 | 中国人民解放军63892部队 | Parameter estimation method of time-frequency aliasing frequency hopping signal |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2007535195A (en) * | 2003-07-11 | 2007-11-29 | ブルービュー テクノロジーズ インコーポレイテッド | Method and system for implementing frequency-steered acoustic arrays for 2D and 3D images |
| US20140297188A1 (en) * | 2013-03-29 | 2014-10-02 | Cgg Services Sa | Time-frequency representations of seismic traces using wigner-ville distributions |
| CN108780538A (en) * | 2016-03-23 | 2018-11-09 | 联邦快递服务公司 | The system, apparatus and method of broadcast setting for the node in self-adjusting wireless node network |
| CN110048741A (en) * | 2019-04-22 | 2019-07-23 | 桂林电子科技大学 | A kind of method for parameter estimation of the Frequency Hopping Signal based on Short-Time Fractional Fourier Transform |
-
2021
- 2021-03-10 CN CN202110258849.9A patent/CN112929053B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102305940A (en) * | 2011-05-24 | 2012-01-04 | 中国石油集团川庆钻探工程有限公司 | Method for extracting fluid factor |
| CN105991492A (en) * | 2015-02-10 | 2016-10-05 | 北京空间飞行器总体设计部 | Frequency hopping (FH) signal identification method |
| CN107276629A (en) * | 2017-06-20 | 2017-10-20 | 中国人民解放军军械工程学院 | Hop cycle method of estimation based on time frequency analysis center of gravity of frequency |
| CN110113075A (en) * | 2018-10-29 | 2019-08-09 | 西安电子科技大学 | Hybrid network platform frequency parameter blind estimating method based on STFT-SPWVD |
| CN111934711A (en) * | 2020-06-30 | 2020-11-13 | 中国人民解放军63892部队 | Parameter estimation method of time-frequency aliasing frequency hopping signal |
Non-Patent Citations (1)
| Title |
|---|
| GMSK跳颇信号参数盲估计和盲检测算法的研究;赵俊;《优秀硕士学位论文全文库》;20030131;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112929053A (en) | 2021-06-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110852201B (en) | Pulse signal detection method based on multi-pulse envelope spectrum matching | |
| CN106597408B (en) | High-order PPS signal parameter estimation method based on time-frequency analysis and instantaneous frequency curve fitting | |
| CN113378661B (en) | Direct-current electric energy signal denoising method based on improved wavelet threshold and related detection | |
| CN107835036B (en) | Non-cooperative frequency hopping signal cracking method | |
| CN103995950A (en) | Wavelet coefficient partial discharge signal noise elimination method based on related space domain correction threshold values | |
| CN110988834B (en) | Pulse arrival time measuring method based on self-adaptive threshold value double thresholds | |
| Le et al. | Weak LFM signal dectection based on wavelet transform modulus maxima denoising and other techniques | |
| CN113472390A (en) | Frequency hopping signal parameter estimation method based on deep learning | |
| CN114785379A (en) | Underwater sound JANUS signal parameter estimation method and system | |
| CN112462321B (en) | Acoustic beacon signal interception processing method based on pulse matching accumulation | |
| CN110542927B (en) | Variable window weighted seismic data spike noise suppression method | |
| CN110632563B (en) | Intra-pulse frequency coding signal parameter measuring method based on short-time Fourier transform | |
| CN112929053B (en) | Frequency hopping signal feature extraction and parameter estimation method | |
| CN110287853B (en) | Transient signal denoising method based on wavelet decomposition | |
| CN112033656A (en) | Mechanical system fault detection method based on broadband spectrum processing | |
| CN116032709B (en) | Method and device for blind demodulation and modulation feature analysis of FSK signal without priori knowledge | |
| CN114650108B (en) | Method and system for detecting signal of transform domain communication system | |
| CN109120562B (en) | MFSK signal frequency estimation method based on spectrum accumulation matching | |
| ZhaoHeng et al. | Selection of the optimal wavelet bases for wavelet de-noising of partial discharge signal | |
| US6853933B2 (en) | Method of identifying spectral impulses for Rj Dj separation | |
| CN115184876A (en) | 2FSK signal parameter estimation method based on wavelet transformation and waveform shaping | |
| CN105959035B (en) | A kind of direct sequence signal intercepts and captures detection method | |
| CN106849992B (en) | A kind of detection method of the direct sequence signal based on Generalized Quadratic power spectrum | |
| Pace et al. | Nyquist folding analog-to-information receiver: Autonomous information recovery using quadrature mirror filtering | |
| Qin et al. | A new method of parameter estimation of frequency-hopping signal |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |