CN106680795A - Time-domain modulation domain parameter combination measure method of frequency agile signal - Google Patents

Abstract

The invention discloses a time-domain modulation-domain parameter joint measurement method of a frequency-agile signal, which mainly solves the problems of low precision and large amount of computation in the prior art for measuring frequency-agile signal parameters. The implementation steps are: 1. Obtain the discrete value of the real part and imaginary part of the frequency-agile signal through a signal analyzer; 2. Calculate the signal envelope of the frequency-agile signal according to the discrete value; 3. Estimate the frequency agility from the signal envelope The approximate amplitude of the signal; 4. Use the approximate amplitude to set the judgment threshold to detect the number of pulses of the frequency-agile signal; 5. Calculate the pulse width of each pulse; 6. Calculate various time-domain parameters from the number of pulses and the pulse width ; 7. Calculate the instantaneous frequency of the frequency-agile signal on the basis of the time-domain parameters; 8. Calculate various modulation domain parameters of the frequency-agile signal from the instantaneous frequency. The method of the invention has the advantages of small calculation amount and higher precision, and can be used for radar signal processing.

Description

Joint Measurement Method of Time-Domain Modulation-Domain Parameters for Frequency-Agile Signals

technical field

The invention relates to the technical field of radar, in particular to a method for joint measurement of time-domain modulation domain parameters of frequency-agile signals, which can be used for radar signal processing.

Background technique

Frequency agile signal refers to a signal type in which the carrier frequency of adjacent pulses of radar signals changes randomly and rapidly within a certain frequency band. Measuring the parameters of frequency-agile signals is one of the focuses in the field of radar signal processing engineering. The traditional measurement method is only for the time domain or the modulation domain, and does not combine the time domain and the modulation domain for parameter measurement, and the measurement of frequency-agile signal modulation domain parameters requires time-frequency analysis. Commonly used time-frequency analysis methods, such as short Time Fourier transform STFT and Wigner-Vile distribution, they have many defects in time-frequency analysis of frequency-agile signals. STFT has the problem of difficult selection of window length and window function. Due to the use of Fourier transform, the amount of calculation is large, and it cannot solve the contradiction between time resolution and frequency resolution. Although the Wigner-Vile distribution does not use a window function, there is no STFT problem, but there are "cross-term" interferences in the time-frequency distribution of multi-component signals, the time-frequency distribution will become blurred, and the frequency estimation is inaccurate, and it The integral operation is adopted, and the calculation amount is relatively large.

Contents of the invention

The purpose of the present invention is to address the deficiencies of the above-mentioned prior art, and propose a method for joint measurement of time-domain modulation domain parameters of frequency-agile signals, so as to reduce the amount of calculation and error, obtain accurate frequency-agile modulation domain parameters, and improve measurement accuracy. precision.

The technical thought of realizing the object of the present invention is joint time domain, modulation domain measurement, adopts the method estimation judgment threshold of statistical pulse signal envelope value during time domain analysis, realizes the measurement of time domain parameters such as pulse number, pulse width, in modulation domain During the analysis, the phase difference method is used to obtain the instantaneous frequency curve of the signal, and to obtain accurate modulation domain parameters such as agile frequency. The technical steps include the following:

(1) Measure the discrete value of the frequency-agile signal;

1a) Read the numerical values of the real part I(t) and the imaginary part Q(t) of the frequency agility signal to be measured with a signal analyzer;

1b) Set the sampling rate f _s , select the real part and imaginary part of the frequency agility signal to be measured with a time length of 1 second, and obtain the discrete value I(n) of the real part and the discrete value of the imaginary part of the frequency agility signal Q(n);

(2) Obtain the time-domain parameters of the frequency-agile signal through discrete values:

(2a) Calculate the frequency-agile signal envelope according to the discrete values I(n) and Q(n) of the real and imaginary parts of the frequency-agile signal:

(2b) Estimate the approximate top value _A'top and the approximate bottom value _{A'ba of the frequency-agile signal according to the frequency-agile signal envelope y(n), and obtain the approximate amplitude measurement value: A'=A'top -} A '_ba;

(2c) Use 10% of the approximate amplitude measurement value 10%A' as the approximate detection threshold of the rising edge and falling edge, detect the pulse number n of the frequency agility signal, and use the approximate detection threshold of the rising edge of each pulse to be located The number of points and the approximate detection threshold of the falling edge are respectively divided by the sampling rate to obtain the approximate judgment time t' of the corresponding _rising edge of each pulse and the approximate judgment time t' of the _falling edge, and calculate each pulse of the pulse signal The pulse width is: τ = t' _rise - t'_fall;

(2d) Within the pulse width of each pulse, use the density distribution average method to calculate the precise top value A _top and bottom value A _ba of each pulse, and obtain the precise amplitude value A of each pulse as: A=A _top -A _ba ;

(2e) Calculate the overshoot S _over and undershoot S _under of each pulse according to the maximum value V _max and the minimum value V _min of each pulse:

S _over =V _max -A _top , S _under =A _ba -V _min ,

(2f) According to the precise top value A _top , bottom value A _ba and precise amplitude value A of each pulse of the frequency-agile signal, calculate the pulse amplitude _upper reference line Mupper and the pulse amplitude lower reference line Mlower _of the frequency-agile signal :

_Above M=A _ba +90%×A, _under M=A _ba +10%×A;

(2g) According to the upper reference line M _of the pulse amplitude, the lower reference line M _{of the} pulse amplitude and the pulse width τ of the frequency agility signal, search for the time value t _{r upper} corresponding to the upper reference line of the pulse amplitude in the rising edge respectively within the pulse width , the time value t _{r lower} corresponding to the pulse amplitude lower reference line and the time value t _{f upper} corresponding to the pulse amplitude upper reference line in the falling edge, the time value t _f lower corresponding to the pulse amplitude lower reference line, and the next rising edge At the moment t' _r corresponding to the upper reference line of the medium pulse amplitude, calculate the rise time t _r , fall time t _f , pulse period T, off time t _off and duty cycle d _t of the frequency agility signal pulse:

t _r =t _{r up-} t _{r down} , t _f =t _{f down} -t _{f up} , T=t' _{r down} -t _{r down} , t _off =T-τ, d _t =τ/T;

(3) Obtain the modulation domain parameters of the frequency-agile signal through discrete values:

(3a) According to the discrete values I(n) and Q(n) of the real and imaginary parts of the frequency-agile signal, calculate the instantaneous phase of the frequency-agile signal:

(3b) According to the instantaneous phase of the frequency-agile signal The time-frequency analysis of the frequency-agile signal is carried out by using the phase difference method, and the instantaneous frequency of the frequency-agile signal is obtained: Where f _s is the sampling rate of the signal;

(3c) Calculate the average value F _avg of the instantaneous frequency of the frequency-agile signal according to the instantaneous frequency f _c (n) of the frequency-agile signal;

(3d) According to the average value F _avg of the instantaneous frequency of the frequency-agile signal and the number of pulses n, count the number N of the instantaneous frequency of the frequency-agile signal near the average value, and if it satisfies 90% n≤N≤n, then judge the The agility mode of the frequency agility signal is intra-pulse agility, otherwise it is judged as inter-pulse agility;

(3e) According to the discrete values I(n) and Q(n) of the real and imaginary parts of the frequency-agile signal, respectively perform fast Fourier transform on the real and imaginary parts of the points in each pulse to obtain the corresponding real Part I(ω) and imaginary part Q(ω), and calculate its instantaneous power:

(3f) According to the instantaneous power spec(ω) of the frequency agility signal, calculate the number C of agility and the frequency freq(i) of the intra-pulse agility or inter-pulse agility, i=1,2,... ,C.

Compared with the prior art, the present invention has the following advantages:

1. In the process of estimating the approximate amplitude of the pulse, the present invention uses the density distribution average method to estimate the approximate top value _A'top and the approximate bottom value _A'ba of the frequency agility signal, compared with the existing density distribution mode method The accuracy is higher. When the number of sampling points of the frequency agile signal is more, the number of intervals equally divided between the maximum value of the pulse envelope y _max and the minimum value of the pulse envelope y _min is more, and the obtained approximate bottom value A' The more accurate _top and the approximate top value A' _ba are, the better accuracy requirements can be met.

2. When the present invention carries out time-frequency analysis, the phase difference method is adopted. Compared with the STFT method, the amount of calculation is greatly reduced, and there is no window function, and is not limited by the uncertainty criterion; compared with the Wigner-Vile distribution method, there is no " Compared with other methods, it has the advantages of small calculation amount of measurement parameters and small error, and is suitable for engineering practice.

The present invention will be described in further detail below in conjunction with the accompanying drawings.

Description of drawings

Fig. 1 is the realization flowchart of the present invention;

Fig. 2 is the time-frequency curve diagram that carries out time-frequency analysis to frequency agile signal with existing STFT method;

Fig. 3 is the time-frequency curve diagram that carries out time-frequency analysis to frequency agile signal with existing Wigner-Vile method;

Fig. 4 is the instantaneous phase curve and time-frequency curve diagram of time-frequency analysis by the method of the present invention.

detailed description

In radar reception, the time domain and modulation domain parameters of the radar echo carry a lot of useful information, but the parameters of the radar echo are unknown and cannot be obtained directly through the radar receiving antenna. For different types of radar signals, the parameter measurement methods are also different. Frequency-agile signal is a common transmission signal of radar. Measuring the parameters of frequency-agile signal is one of the key points in the field of radar signal processing engineering. Due to the complex characteristics of intra-pulse modulation, the frequency-agile signal makes parameter measurement particularly difficult, and the amount of calculation is huge, and the problem of inaccurate measurement often occurs. In order to accurately and quickly measure the time-domain and modulation-domain parameters of frequency-agile signals, a measurement method with a small amount of calculation and higher precision must be adopted. The present invention reads the numerical value of the frequency agility signal through the signal analyzer, and then adopts the idea of combining the time domain and the modulation domain to measure the specific parameters of the frequency agility signal more accurately and in detail by using the density distribution averaging method and the phase difference method.

With reference to Fig. 1, the implementation steps of the present invention are as follows:

Step 1: Measure the discrete values of the frequency-agile signal.

1a) The radar receives the radar echo in the form of a frequency-agile signal through the receiving antenna, and saves the real part I(t) and imaginary part Q(t) of the received frequency-agile signal in a .dat file in the form of data;

1b) Upload the .dat file to the signal analyzer, and read the values of the real part I(t) and the imaginary part Q(t) of the frequency-agile signal;

1c) Set the sampling rate f _s , extract the frequency-agile signal to be measured with a length of 1 second, and perform discrete sampling on it to obtain the real part discrete value I(n) and The imaginary part discretely takes the value Q(n).

Step 2: Calculate the time-domain parameters of the frequency-agile signal.

2a) Calculate the frequency-agile signal envelope according to the discrete values I(n) and Q(n) of the real and imaginary parts of the frequency-agile signal:

2b) Estimating the approximate top value A' _top and the approximate bottom value A' _ba of the frequency-agile signal, as well as the approximate amplitude A';

2b1) Find the maximum value y _max and the minimum value y _min of the signal envelope y(n), where n=1, 2, ..., N, N is the total number of points obtained by sampling the frequency agility signal;

2b2) Divide the signal envelope y(n) into 100 intervals from small to large, where the minimum value of the k-th interval E(k) is E _min (k), and the maximum value is E _max (k):

2b3) Statistically count the number of signal envelopes y(n) falling in each interval E(k), which are respectively recorded as count values C(k), k=1, 2...100;

2b4) Find the position of the maximum value of the first 50 count values in the count value C(k), denoted as k ₁ , and find the position of the maximum value of the last 50 count values, denoted as k ₂ ;

2b5) Calculate the average value of the interval E(k ₁ ), which is the approximate top value A' _top of the frequency-agile signal, and calculate the average value of the interval E(k ₂ ), which is the approximate bottom value A' of the frequency-agile signal _ba ;

2b6) According to the approximate bottom value _A'ba and the approximate top value _A'top of the frequency-agile signal, the approximate amplitude A'= _A'top - _A'ba of the frequency-agile signal is obtained;

2c) Detecting the number m of pulses of the frequency agility signal, and recording the approximate _rising edge judgment time t' of each pulse, and the approximate _falling edge judgment time t'fall of each pulse;

2c1) Record the number of pulses of the frequency agility signal as m, and initialize m to 0, start to search for the first approximate rising edge from the initial position of the frequency agility signal, and record the search result as the flag bit, if the search If successful, set the flag bit to 1, and record the approximate _rising edge judgment time t', otherwise, set the flag bit to 0;

2c2) After finding the approximate rising edge, continue to search for the approximate falling edge. If the search is successful, set the flag bit flag to 1, and record the judgment time t'fall of the approximate _falling edge, otherwise, set the flag bit flag to 0 ;

2c3) After finding the approximate falling edge, continue to search for the approximate rising edge. If the search is successful, set the flag bit flag to 1, record the judgment time t' of the approximate _rising edge, and make the pulse number of the frequency agility signal Add 1 to the number m, otherwise, set the flag bit to 0;

2c4) detect the current value of the flag bit flag, if the flag bit flag=1, then return to step 2c2), if the flag bit flag=0, then end the search;

2d) Calculate the width of each pulse of the frequency agility signal: τ=t'down _{- t'up} ;

2e) Within the pulse width of each pulse, use the density distribution averaging method to calculate the precise top value A _top and bottom value A _ba of each pulse, and obtain the precise amplitude value A of each pulse as: A=A _top -A _ba ;

2f) Detecting the maximum value V _max and the minimum value V _min of each pulse of the frequency agility signal, and calculating the overshoot S _over =V _max -A _top and the undershoot S _under =A _ba -V _min of each pulse;

2g) Calculate the pulse rise time t _r , fall time t _f , pulse period T, off time t _off and duty cycle d _t :

2g1) Set the upper reference line M _{of the} pulse amplitude and the lower reference line M _of the pulse amplitude of the frequency agility signal:

M _up =A _ba +90%×A,

_Under M=A _ba +10%×A;

2g2) Find the corresponding time values _on the pulse amplitude upper reference line M and pulse amplitude _lower reference line M within the pulse width τ. The corresponding time values include: the time t _r corresponding to the pulse amplitude upper reference line in the rising edge , the time t _{r down} corresponding to the lower reference line of the pulse amplitude in the rising edge, the time t _f corresponding to the upper reference line of the pulse amplitude in the falling edge, and the time t _{f lower} corresponding to the lower reference line of the pulse amplitude in the falling edge, and The time _t'r corresponding to the upper reference line of the pulse amplitude in the adjacent next rising edge;

2g3) Calculate the following parameters of the pulse according to the moment value found in 2g2):

Rise time t _r : t _r = t _{r up -} t _{r down} ,

Falling time t _f : t _f = t _{f down -} t _{f up} ,

Pulse period T: T = t' _{r down -} t _{r down} ,

Off time t _off : t _off =T-τ,

Duty ratio d _t : d _t =τ/T.

Step 3: Calculate the modulation domain parameters of the frequency-agile signal.

3a) According to the discrete values I(n) and Q(n) of the real and imaginary parts of the frequency-agile signal, calculate the instantaneous phase of each point of the frequency-agile signal: n=1,2,...,N, where N is the total number of points sampled by the frequency agility signal;

3b) According to the instantaneous phase The time-frequency analysis of the frequency-agile signal is carried out by using the phase difference method, and the instantaneous frequency of the frequency-agile signal is extracted: n=1,2,...,N, where f _s is the sampling rate of the signal, and N is the total number of points sampled by the frequency agility signal;

3c) According to the instantaneous frequency f _c (n) of each point of the frequency agile signal, obtain the average value of the instantaneous frequency of the frequency agile signal;

3d) Count the number M of the instantaneous frequency of the frequency-agile signal near the average value. If 90% m≤M≤m is satisfied, the agile mode of the frequency-agile signal is judged to be intra-pulse frequency-agile, otherwise it is judged to be pulse-agile. Time-agile frequency, m is the number of pulses of the frequency-agile signal;

3e) According to the discrete values I(n) and Q(n) of the real and imaginary parts of the frequency-agile signal, fast Fourier transform is performed on the real and imaginary parts of the points in each pulse respectively to obtain the corresponding real part I(ω) and imaginary part Q(ω), and calculate its instantaneous power:

3f) According to the instantaneous power spec(ω) of the frequency agility signal, calculate the number C of agility in the pulse agility mode and the agility frequency freq(i), i=1,2,...,C;

3f1) Statistically obtain the value range spec(j) of the instantaneous power spec(ω), j=1,2,...,N, where N is the total number of points sampled by the frequency agility signal;

3f2) Calculate the instantaneous mean value avg: avg=(spec(1)+spec(2)+...+spec(j)+...+spec(N))/N;

3f3) Judge the value of each range spec(j) in turn, if it satisfies spec(j)>avg and spec(j-2)<spec(j-1), spec(j-1)<spec(j ), spec(j)>spec(j+1), spec(j+1)>spec(j+2), then record this point as a spectral peak pSpec(ω);

3f4) Statistically find all the spectrum peaks pSpec(ω), compare the size of each spectrum peak pSpec(ω) in turn, determine the maximum spectrum peak pSpec _max (ω) and the number of maximum spectrum peaks, the maximum spectrum peak number The number is the agile number C;

3f5) Divide the pulse width of each pulse into agile segments, and calculate the fitting frequency for each segment respectively, which is the agile frequency freq(i), i=1, 2,...,C;

3g) According to the instantaneous power spec(ω) of the frequency agility signal, calculate the number of agility C and the agility frequency freq(i) of the pulse-to-pulse agility mode, i=1,2,...,C;

3g1) Calculate the fitting frequency freq(k) for each pulse of the frequency-agile signal, k=1,2,...,m, wherein m is the number of pulses of the frequency-agile signal;

3g2) Find the maximum value freq _max and the position of the pulse in the fitted frequency freq(k);

3g3) According to the pulse position where the maximum value freq _max of the frequency agility signal fitting frequency is located, calculate the difference between the pulse positions where two adjacent maximum values are located, and the difference is the number C of the agility;

3g4) Take the fitting frequency of the previous number C of agile changes, which is the agile frequency freq(i), i=1,2,...,C.

The effect of the present invention on the time domain and modulation domain parameter measurement of the frequency agile signal can be further verified by the following simulation experiments.

Experiment 1, carry out Matlab emulation to the time-domain parameter measurement of frequency agile signal with the inventive method, parameter is set as: agile mode is pulse-to-pulse agility, and pulse width is 10us, and pulse repetition period is 60us, and the complete pulse number is 6. The sampling rate is 20MHz and the bandwidth is 20MHz. The measured time domain parameters are shown in Table 1:

Table 1 Time Domain Parameters of Frequency Agility Signal

It can be seen from Table 1 that the number of measured pulses is 6, the pulse width is 10us, the repetition period is 60us, the rise time is 0.05us, and the fall time is 0.05us. The measurement parameters are consistent with the actual set parameters.

Experiment 2, using the method of the present invention to carry out Matlab simulation on the modulation domain parameter measurement of the frequency agility signal, the parameters are set to: bandwidth 20Mhz, maximum frequency and minimum frequency in the pulse are 10.2e+008 and 9.8e+008. The measured pulse modulation parameters are shown in Table 2:

Table 2 Frequency-agile signal intra-pulse modulation parameter list

It can be seen from Table 2 that the agile mode is pulse-to-pulse agility, the number of agility is 3, and the agility frequencies are 0MHz, 1.5MHz, and 3MHz respectively. The measurement results are accurate and the effect is very good.

In Experiment 3, the existing STFT method and the Wigner-Vile method were used to analyze the time-frequency of the frequency-agile signal to obtain the corresponding time-frequency curve. The measurement results are shown in Figure 2 and Figure 3. 2 is a time-frequency distribution diagram of a frequency-agile signal measured by the STFT method, and FIG. 3 is a time-frequency distribution diagram of a frequency-agile signal measured by the Wigner-Vile method.

It can be seen from Fig. 2 that the resolution of the time-frequency distribution map obtained by the existing STFT method is relatively poor.

It can be seen from Figure 3 that with the existing Wigner-Vile distribution method, the time-frequency distribution of the frequency-agile signal appears "false", and the useful time-varying spectrogram becomes blurred, so the STFT and Wigner-Vile distributions are not Ideal for analyzing frequency-agile signals.

Experiment 4, with the method of the present invention, frequency-agile signal is carried out time-frequency analysis to obtain instantaneous phase curve and time-frequency curve, measurement result as shown in Figure 4, wherein Figure 4 (a) is the instantaneous phase curve of frequency-agile signal, Figure 4(b) is the time-frequency curve of the frequency-agile signal, Figure 4(c) is the enlarged picture of the instantaneous phase curve of the frequency-agile signal at a frequency of 1.5MHz, and Figure 4(d) is the frequency-agile signal Zoom-in view of the instantaneous phase curve at a frequency of 3MHz.

It can be seen from the simulation results in Fig. 4 that the method for measuring frequency-agile signal parameters of the present invention has the characteristics of simple algorithm, small amount of calculation, etc., and there is no "cross-term" interference, which is especially suitable for large data volume in engineering practice. Measurement.

To sum up, the method for measuring frequency-agile signal parameters of the present invention can more accurately perform time-frequency analysis on the frequency-agile signal, and obtain various time-domain parameters and modulation-domain parameters more quickly.

Claims (5)

1. A time-domain modulation domain parameter joint measurement method of a frequency-agile signal, comprising: (1) Measure the discrete value of the frequency-agile signal; (1a) Read the numerical value of the real part I(t) and the imaginary part Q(t) of the frequency agility signal to be measured with a signal analyzer; (1b) Set the sampling rate f _s , sample the real part and imaginary part of the frequency-agile signal to be measured with a time length of 1 second, and obtain the discrete value of the real part I(n) and the discrete value of the imaginary part of the frequency-agile signal Value Q(n); (2) Obtain the time-domain parameters of the frequency-agile signal through discrete values: (2a) Calculate the frequency-agile signal envelope according to the discrete value I(n) of the real part and the discrete value Q(n) of the imaginary part of the frequency-agile signal: (2b) Estimate the approximate top value _A'top and the approximate bottom value _{A'ba of the frequency-agile signal according to the frequency-agile signal envelope y(n), and obtain the approximate amplitude measurement value: A'=A'top -} A '_ba; (2c) Use 10% of the approximate amplitude measurement value 10%A' as the approximate detection threshold of the rising edge and falling edge, detect the pulse number m of the frequency agility signal, and use the approximate detection threshold of the rising edge of each pulse The number of points and the number of points where the approximate detection threshold of the falling edge is divided by the sampling rate respectively, and the approximate judgment time t' of the corresponding _rising edge of each pulse and the approximate judgment time t' of the _falling edge of each pulse are obtained, and the calculation of each pulse signal The pulse width of the pulse is: τ=t'fall _{- t'rise} ; (2d) Within the pulse width of each pulse, use the density distribution average method to calculate the precise top value A _top and bottom value A _ba of each pulse of the frequency-agile signal, and obtain the precise value of each frequency-agile signal Amplitude value A is: A=A _top -A _ba ; (2e) Calculate the overshoot value S _over and the undershoot value S _under of each pulse according to the maximum value V _max and the minimum value V _min of each pulse of the frequency agile signal: S _over = V _max - A _top , S _under = A _ba - V _min ; (2f) According to the precise top value A _top , bottom value A _ba and precise amplitude value A of each pulse of the frequency-agile signal, calculate the pulse amplitude _upper reference line Mupper and the pulse amplitude lower reference line Mlower _of the frequency-agile signal : _Above M=A _ba +90%×A, _under M=A _ba +10%×A; (2g) According to the upper reference line M _of the pulse amplitude, the lower reference line M _{of the} pulse amplitude and the pulse width τ of the frequency agility signal, search for the time value t _{r upper} corresponding to the upper reference line of the pulse amplitude in the rising edge respectively within the pulse width , the time value t _{r lower} corresponding to the pulse amplitude lower reference line and the time value t _{f upper} corresponding to the pulse amplitude upper reference line in the falling edge, the time value t _f lower corresponding to the pulse amplitude lower reference line, and the next rising edge At the moment t' _r corresponding to the upper reference line of the medium pulse amplitude, calculate the rise time t _r , fall time t _f , pulse period T, off time t _off and duty cycle d _t of the frequency agility signal pulse: t _r =t _{r up-} t _{r down} , t _f =t _{f down} -t _{f up} , T=t' _{r down} -t _{r down} , t _off =T-τ, d _t =τ/T; (3) Obtain the modulation domain parameters of the frequency-agile signal through discrete values: (3a) According to the discrete values I(n) and Q(n) of the real and imaginary parts of the frequency-agile signal, calculate the instantaneous phase of the frequency-agile signal: (3b) According to the instantaneous phase of the frequency-agile signal The time-frequency analysis of the frequency-agile signal is carried out by using the phase difference method, and the instantaneous frequency of the frequency-agile signal is obtained: Where f _s is the sampling rate of the signal; (3c) Calculate the average value F _avg of the instantaneous frequency of the frequency-agile signal according to the instantaneous frequency f _c (n) of the frequency-agile signal; (3d) According to the average value F _avg of the instantaneous frequency of the frequency-agile signal and the number of pulses m, count the number M of the instantaneous frequency of the frequency-agile signal near the average value, and if it satisfies 90% m≤M≤m, then judge the The agility mode of the frequency agility signal is intra-pulse agility, otherwise it is judged as inter-pulse agility; (3e) According to the discrete values I(n) and Q(n) of the real and imaginary parts of the frequency-agile signal, respectively perform fast Fourier transform on the real and imaginary parts of the points in each pulse to obtain the corresponding real Part I(ω) and imaginary part Q(ω), and calculate its instantaneous power: (3f) According to the instantaneous power spec(ω) of the frequency agility signal, calculate the number C of agility and the frequency freq(i) of the intra-pulse agility or inter-pulse agility, i=1,2,... ,C. 2. The method according to claim 1, wherein the approximate top value A' _top and the approximate bottom value A' _ba of the pulse are estimated according to the signal envelope y (n) in the step (2b), as follows: 2b1) Find the maximum value y _max and the minimum value y _min in the signal envelope y(n), where n=1, 2,..., N, N is the total number of points obtained by sampling the frequency agility signal; 2b2) Divide the signal envelope y(n) into 100 intervals from small to large, and calculate the minimum value E _min (k) and maximum value E _max (k) of the value range of the k-th interval E(k): ${\mathrm{<span\; class="notranslate">\; E.</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}\mathrm{<span\; class="notranslate">\; i</span>}\mathrm{<span\; class="notranslate">\; no</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}\mathrm{<span\; class="notranslate">\; i</span>}\mathrm{<span\; class="notranslate">\; no</span>}}<span\; class="notranslate">\; +</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span>\frac{{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}\mathrm{<span\; class="notranslate">\; a</span>}\mathrm{<span\; class="notranslate">\; x</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; min</span>}}}{\mathrm{<span\; class="notranslate">\; 100</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; E.</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}\mathrm{<span\; class="notranslate">\; a</span>}\mathrm{<span\; class="notranslate">\; x</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; min</span>}}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; k</span>}\frac{{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}\mathrm{<span\; class="notranslate">\; a</span>}\mathrm{<span\; class="notranslate">\; x</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; min</span>}}}{\mathrm{<span\; class="notranslate">\; 100</span>}}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; 2</span>}<span\; class="notranslate">\; ,</span><span\; class="notranslate">\; ...</span><span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; 100</span>}<span\; class="notranslate">\; ;</span>$ 2b3) Statistically count the number of signal envelopes y(n) falling in each interval E(k), which are respectively recorded as count values C(k), k=1, 2...100; 2b4) Find the position of the maximum value of the first 50 count values in the count value C(k), denoted as k ₁ , and find the position of the maximum value of the last 50 count values, denoted as k ₂ ; 2b5) Calculate the average value of the interval E(k ₁ ), which is the approximate bottom value A' _ba of the pulse, and calculate the average value of the interval E(k ₂ ), which is the approximate top value A' _top of the pulse. 3. method according to claim 1, wherein in said step (2c) with 10%A ' as rising edge, falling edge detection threshold, measure the pulse number of frequency agility signal, carry out as follows: 2c1) Record the number of pulses of the frequency agility signal as m, and initialize m to 0, start to search for the first approximate rising edge from the initial position of the frequency agility signal, and record the search result as the flag bit, if the search If successful, set the flag bit to 1, otherwise, set the flag bit to 0; 2c2) After the approximate rising edge is found, continue to search for the approximate falling edge. If the search is successful, the flag bit flag is set to 1, otherwise, the flag bit flag is set to 0; 2c3) After the approximate falling edge is found, continue to search for the approximate rising edge. If the search is successful, set the flag bit flag to 1, and increase the pulse number m of the frequency agility signal by 1, otherwise, set the flag bit flag to 0; 2c4) Detect the current value of the flag bit, if the flag bit flag=1, return to step 2c2), if the flag bit flag=0, end the search. 4. The method according to claim 1, wherein in the step (3f), calculate the agile number C and the agile frequency freq(i) of the intrapulse agile mode, as follows: 3f1) Statistically obtain the value range spec(j) of the instantaneous power spec(ω) of the frequency-agile signal, j=1,2,...,N, wherein N is the total points obtained by sampling the frequency-agile signal; 3f2) Calculate the instantaneous mean value: avg=(spec(1)+spec(2)+...+spec(j)+...+spec(N))/N; 3f3) Judge the value of each range spec(j) in turn, if it satisfies spec(j)>avg and spec(j-2)<spec(j-1), spec(j-1)<spec(j ), spec(j)>spec(j+1), spec(j+1)>spec(j+2), then record this point as a spectral peak pSpec(ω); 3f4) Statistically find all the spectrum peaks pSpec(ω), compare the size of each spectrum peak pSpec(ω) in turn, determine the maximum spectrum peak pSpec _max (ω) and the number of maximum spectrum peaks, the maximum spectrum peak number The number is the agile number C; 3f5) Divide the pulse width τ of each pulse into agility segments, and calculate the fitting frequency for each segment respectively, which is the agility frequency freq(i), i=1,2,...,C. 5. The method according to claim 1, wherein in the step (3f), calculate the agile number C and the agile frequency freq (i) of the pulse-to-pulse agility mode, as follows: 3f6) Calculate the fitting frequency freq(k) for each pulse of the frequency-agile signal, k=1, 2,..., m, wherein m is the number of pulses of the frequency-agile signal; 3f7) Find the maximum value freq _max and the position of the pulse in the fitted frequency freq(k); 3f8) According to the pulse position where the maximum value freq _max of the frequency agility signal fitting frequency is located, calculate the difference between the pulse positions where two adjacent maximum values are located, and the difference is the number C of the agility; 3f9) Take the fitting frequency of the previous number C of agile changes, which is the agile frequency freq(i), i=1,2,...,C.