CN110336587B - Method for acquiring combined time-frequency distribution in multi-frequency-hopping signal reconnaissance - Google Patents
Method for acquiring combined time-frequency distribution in multi-frequency-hopping signal reconnaissance Download PDFInfo
- Publication number
- CN110336587B CN110336587B CN201910640977.2A CN201910640977A CN110336587B CN 110336587 B CN110336587 B CN 110336587B CN 201910640977 A CN201910640977 A CN 201910640977A CN 110336587 B CN110336587 B CN 110336587B
- Authority
- CN
- China
- Prior art keywords
- time
- frequency
- window
- combined
- distribution
- 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
- H04B17/00—Monitoring; Testing
- H04B17/30—Monitoring; Testing of propagation channels
- H04B17/391—Modelling the propagation channel
- H04B17/3912—Simulation models, e.g. distribution of spectral power density or received signal strength indicator [RSSI] for a given geographic region
Abstract
The invention discloses a method for acquiring combined time-frequency distribution in multi-hop signal reconnaissance, which is characterized in that Spectrogram processing is improved on the basis of the conventional SP & SPWVD, firstly, a Hamming window is selected as a window function to carry out long-window and short-window short-time Fourier transform, then, the long-window short-time Fourier transform with higher frequency resolution and a time-frequency transform matrix of the short-window short-time Fourier transform with higher time resolution are subjected to modulo multiplication to obtain a time-frequency matrix SP (m, n) of a combined window Spectrogram (SP, n), then, the time-frequency matrix of the combined window Spectrogram is subjected to truncation processing to obtain a time-frequency matrix SP '(m, n), and finally, the time-frequency matrix SP' (m, n) and the SPWVD are subjected to Hadamard product to obtain combined time-frequency distribution (namely, the improved SP & SPWVD combined time-frequency distribution). Compared with the existing SP & SPWVD combined time-frequency distribution, the time-frequency resolution is further improved, the time-frequency graph is clearer and more stable, and the calculated amount is not increased.
Description
Technical Field
The invention belongs to the technical field of time frequency analysis, and particularly relates to a method for acquiring combined time frequency distribution in multi-frequency-hopping signal reconnaissance.
Background
Because the frequency hopping signal has excellent anti-interference performance, lower interception probability and stronger multiple access networking capability, the frequency hopping communication technology is widely applied to the fields of military, civil use and the like in recent years. Meanwhile, the investigation of the frequency hopping signal becomes an important research content, and the analysis of the frequency hopping signal is essential in the investigation of the frequency hopping signal. The frequency of the frequency hopping signal changes constantly with time, and belongs to typical non-stationary signals, which are difficult to be accurately analyzed by a simple time domain or frequency domain analysis method, and the time frequency analysis technology is a very effective method for processing the non-stationary signals. The quality of the time-frequency analysis result can directly influence the reconnaissance result of the whole frequency hopping signal.
Time-frequency analysis methods for frequency hopping signals are generally classified into linear time-frequency distribution and nonlinear time-frequency distribution. The short-time Fourier transform is classical linear time-frequency distribution, is windowed Fourier transform, has the advantages of low algorithm complexity, high operation speed by using fast Fourier transform, no cross term interference and low time-frequency resolution, and cannot meet the application scene with high time-frequency resolution requirement due to the mutual restriction of the time resolution and the frequency resolution. The spectrogram is the square of the short-time fourier transform, which is theoretically a non-linear time-frequency distribution, but it has no cross terms when the signal components do not overlap, although the amount of computation is increased with respect to the short-time fourier transform, while the frequency resolution is improved. The wigner-weil distribution (WVD) is typically a non-linear time-frequency distribution with the best time-frequency resolution, but also with severe cross-term interference. The smooth pseudo-wigner distribution (SPWVD) is generated by simultaneously adding a time domain window and a frequency domain window on the basis of the wigner-wiener distribution (WVD). Its advantages are suppressing the interference of Wegener-Weill distribution (WVD) and high computation load and low time-frequency resolution.
In various time-frequency distributions, time-frequency focusing and cross item interference are always a pair of contradictions, and the time-frequency distribution with WVD time-frequency focusing and no cross item interference does not exist. Therefore, some scholars think that the time frequency distribution with better time frequency focusing performance and the time frequency distribution without cross item interference are combined to achieve satisfactory comprehensive effect.
In the paper "time-frequency analysis [ J ] of frequency hopping signals" (aerospace, 2009,30(2): 740) 747) of Chenlihu, Zhang and Shenrong, the time-frequency analysis effects of different time-frequency distributions on the simultaneous existence of multi-frequency signals are comprehensively compared, and clear conclusion of a time-frequency graph can be obtained by a spectrogram, smooth pseudo-Wegener distribution (SPWVD), rearrangement type time-frequency distribution and a linear and nonlinear time-frequency distribution combination method, but the time-frequency resolution of the combination distribution time-frequency graph provided in the text is still to be improved.
In the application [ J ] "(the university of electronics and technology, 2010(6): 1137-.
In a thesis "morphological filtering and combined time-frequency distribution frequency hopping signal parameter estimation [ J ]" (thz science and electronic information science, 2013(6)) in Zhao Fangji, Jiang, Guo military and the like, a frequency hopping parameter blind estimation method based on morphological filtering and combined time-frequency distribution is provided. Compared with the method for estimating the frequency hopping parameters by directly utilizing smooth pseudo-Wigner (SPWVD), the method has the advantages of smaller calculated amount and higher estimation accuracy, but is not compared with other combined time frequency distribution.
In a thesis of Chua-Egyu and Chen Yingzhi, namely 'frequency hopping signal analysis [ J ]' based on combined time-frequency distribution (experimental science and technology, 2014,12(04):35-36+117), a spectrogram and SPWVD combined time-frequency distribution are used for frequency hopping signal analysis, the period of a single frequency hopping signal is effectively estimated, and a scene in which a plurality of frequency hopping signals exist simultaneously is not verified.
In a frequency hopping signal period blind estimation algorithm [ J ] "(computer engineering and design, 2016,37(11):2861 + 2964) in a paper of Tangning, Guo English, Zhang Kunmeng, Zhang Dongwei and Yujian military, a frequency hopping signal period blind estimation algorithm based on Gabor and SPWVD combined time-frequency distribution is provided, and the algorithm has strong signal-to-noise ratio adaptability, high parameter estimation precision, but is complex and large in calculation amount.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provide a method for acquiring combined time-frequency distribution in multi-frequency signal reconnaissance, which further improves the time-frequency resolution and obtains a clearer and more stable time-frequency graph on the basis of SP & SPWVD combined time-frequency distribution without increasing the calculated amount of the SP & SPWVD combined time-frequency distribution so as to meet the reconnaissance requirement of the multi-frequency signal.
In order to achieve the above object, the method for obtaining combined time-frequency distribution in multi-frequency-hopping signal reconnaissance of the present invention is characterized by comprising the following steps:
(1) determining an analysis signal z (t) of the multi-frequency-hopping signal y (t);
(2) and performing long-window short-time Fourier transform on the analytic signal z (t) to obtain a time-frequency transform matrix STFTh1(m, n), wherein m is a time point, n is a discrete frequency point, and the window function is a Hamming window;
(3) and carrying out short-window short-time Fourier transform on the analytic signal z (t) to obtain a time-frequency transform matrix STFTh2(m, n), the window function being a hamming window;
(4) transforming the time-frequency matrix STFTh1(m,n)、STFTh2(m, n) multiplying after modulus calculation to obtain a time frequency matrix SP (m, n) of the combined window spectrogram;
(5) and cutting off the time frequency matrix SP (m, n) of the combined window spectrogram to obtain a time frequency matrix SP' (m, n) of the combined window spectrogram:
wherein the truncation threshold epsilon is:
wherein eta is a threshold factor and can be adjusted according to the signal-to-noise ratio, M is the number of time sampling points, and N is the number of frequency points;
(6) solving a smooth pseudo Wigner distribution time-frequency matrix SPWVD (m, n) of the analytic signal z (t);
(7) and performing Hadamard product on the time-frequency matrix SP' (m, n) of the truncated combined window spectrogram and the SPWVD time-frequency matrix SPWVD (m, n) to obtain combined time-frequency distribution (namely improved SP & SPWVD combined time-frequency distribution).
The object of the invention is thus achieved.
The invention discloses a method for acquiring combined time-frequency distribution in multi-hop signal reconnaissance, which is characterized in that Spectrogram processing is improved on the basis of the conventional SP & SPWVD, firstly, a Hamming window is selected as a window function to carry out long-window and short-window short-time Fourier transform, then, a long-window short-time Fourier transform with higher frequency resolution and a time-frequency transform matrix of a short-window short-time Fourier transform with higher time resolution are subjected to modulo multiplication to obtain a time-frequency matrix SP (m, n) of a combined window Spectrogram (SP, i.e. Spectrogram) and then, a time-frequency matrix of the combined window Spectrogram is subjected to truncation processing to obtain a time-frequency matrix SP '(m, n), and finally, the time-frequency matrix SP' (m, n) and the SPWVD are subjected to Hadamard product to obtain combined time-frequency distribution (namely, the improved SP & SPWVD combined time-frequency distribution). Compared with the existing SP & SPWVD combined time-frequency distribution, the time-frequency resolution is further improved, the time-frequency graph is clearer and more stable, and the calculated amount is not increased.
Drawings
FIG. 1 is a flowchart of an embodiment of a method for obtaining combined time-frequency distribution in multi-frequency signal reconnaissance according to the present invention;
FIG. 2 is a STFT & WVD combined time-frequency distribution plot;
FIG. 3 is a STFT & PWVD combined time-frequency distribution plot;
FIG. 4 is a STFT & SPWVD combined time-frequency distribution plot;
FIG. 5 is a SP & SPWVD combined time-frequency distribution plot;
FIG. 6 is a diagram of a combined time-frequency distribution obtained by the present invention.
Detailed Description
The following description of the embodiments of the present invention is provided in order to better understand the present invention for those skilled in the art with reference to the accompanying drawings. It is to be expressly noted that in the following description, a detailed description of known functions and designs will be omitted when it may obscure the subject matter of the present invention.
1. Frequency hopping signal model
The carrier frequency of the frequency hopping signal is controlled by the pseudo random sequence, and the output frequency continuously hops along with time. A typical frequency hopping signal appears as a short horizontal line of uniform length on a time-frequency diagram, i.e., one frequency is maintained for a period of time, and the next period hops randomly to another frequency and remains for the same time. In the analysis process, signal parameters such as the frequency hopping frequency, the hopping moment, the frequency hopping period and the like of the frequency hopping signal are mainly concerned, and the generation process of the frequency hopping signal is not concerned.
Assuming that a receiver receives a certain frequency hopping signal, the observation time is delta t, wherein the first hop and the last hop are incomplete hops, and the duration of the first received incomplete hop is delta tsCarrier frequency of fstart(ii) a The last incomplete jump having duration Δ teCarrier frequency of fend. In addition, the method comprises K complete hops, and the duration of each hop is ThI.e. with a frequency hopping period of ThThe carrier frequency corresponding to the kth complete hop is fk. The mathematical model of the frequency hopping signal is simplified as:
wherein the content of the first and second substances,
1.1 Multi-frequency hopping Signal model
With the increasing complexity of electromagnetic environment, the receiver receives more than one frequency hopping signal in reality, and the frequency hopping signal is a mixed signal of a plurality of frequency hopping signals. Under the condition of adopting single antenna to receive, assuming that N frequency hopping signals are received simultaneously and all noise interference is additive noise, the mathematical model of the multi-frequency hopping signals is as follows:
1.2 Multi-hop frequency Signal STFT and spectrogram
The basic idea of short-time fourier transform (STFT) is: a time-limited window function is multiplied before Fourier transform of the signal, and the non-stationary signal is assumed to be stationary in a short time interval of an analysis window, and the signal enters an analyzed state section by section through moving the window on a time axis, so that a group of local frequency spectrums of the signal can be obtained, and the time-varying characteristic of the signal can be obtained from the difference of the local frequency spectrums at different moments. In short, the short-time fourier transform (STFT) is a windowed Fourier Transform (FT), so the short-time fourier transform (STFT) of the frequency hopping signal x (t) is defined as:
the discrete form is:
for multiple frequency-hopping signals, e.g. when s (t) x1(t)+x2(t), its short-time fourier transform can be expressed as:
as can be seen from the above formula, the fourier transform of the plurality of frequency hopping signals is the fourier transform addition of each frequency hopping signal.
The Spectrum (SP) is the square of the short-time fourier transform mode, so the spectrum of signal x (t) is defined as:
similarly, for multiple hopping signals, when s (t) x1(t)+x2(t), the spectrogram can be represented as:
wherein the content of the first and second substances,
for frequency-hopping signals x1(t) a spectrum of the light,
for frequency-hopping signals x2(t) a spectrum of the light,
is the cross term of the spectrogram of the two frequency hopping signals,
and
respectively, the phases of the short-time fourier transforms of the two hopping signals. Due to the nature of Fourier transform, the energy-limited signal is decomposed into ejωtSpatially on an orthogonal basis, so as long as the short-time Fourier transforms of the sub-signals do not overlap in the time-frequency domain, the phases of the sub-signals are orthogonal, i.e.
Equal to zero, no cross terms of the spectrogram are present.
1.3 Multi-frequency hopping signals WVD and SPWVD
Wigner distribution (WVD), which was originally applied to the research of quantum mechanics as proposed by Wigner in 1932, was introduced by Vile into the field of signal analysis in 1948, forming the well-known Wigner distribution (WVD). The wigner distribution is essentially the fourier transform of the instantaneous correlation function of the signal with respect to the variable delay, and therefore, for a single frequency hopping signal x (t), its wigner distribution (WVD) is defined as:
when there are multiple frequency hopping signals, e.g. s (t) ═ x1(t)+x2(t), the wigner distribution (WVD) can be expressed as:
wherein
For frequency-hopping signals x1(t) a Weignenwell distribution,
for frequency-hopping signals x2(t) a Weignenwell distribution,
is the cross term of the two frequency hopping signals. The frequency hopping signal itself is a multi-component signal that, according to the convolution theorem,
cross terms may also occur in (1). Therefore, for multi-frequency hopping signals, the cross terms of the wigner distribution (WVD) are very serious.
The smooth wigner distribution (SPWVD) is generated to suppress the cross terms of the wigner distribution. The method simultaneously inhibits cross terms from two directions of time and frequency, namely, a time domain smoothing window and a frequency domain smoothing window are simultaneously added to the Weigner distribution, and the expression is as follows:
smoothing the wigner distribution (SPWVD) greatly reduces cross term interference of the wigner distribution, but reduces time-frequency resolution and increases computational effort.
2. Improvement based on SP & SPWVD combined time frequency distribution
2.1 selection of Combined time-frequency distribution
As can be seen from the analysis in section 1, focality and cross term interference are always a pair of contradictions in various time-frequency distributions, and it has been demonstrated that a time-frequency distribution without cross term interference and with WVD focality is not present. Although finding a distribution which is totally beautiful is impossible, the different frequency distributions can be combined, and the good characteristics of the distributions are utilized to obtain a satisfactory result through comprehensive analysis. The principle of the combined distribution is that a time-frequency distribution function without cross terms is firstly cut off according to a threshold to obtain an effective area of a signal on a time-frequency map, and then the cut-off signal time-frequency map and the distribution function with better time-frequency focusing are subjected to 'multiplication' (Hadamard product), so that the combined time-frequency distribution is obtained. The combined distribution not only eliminates the interference of most cross terms, but also retains the time-frequency focusing performance of signal self terms. The combination of different time-frequency distributions has different effects, and the best time-frequency diagram result can be obtained only by selecting the most appropriate combination mode.
For short-time fourier transforms and spectra, it is not difficult to derive: in a time frequency matrix obtained by the short-time Fourier transform of the signal, the value of the time frequency matrix corresponding to the time frequency point with the signal is larger, and the value of the time frequency matrix corresponding to the part without the signal is smaller. And the spectrogram is the square of short-time Fourier transform, so in a spectrogram time-frequency matrix obtained after the time-frequency matrix of the short-time Fourier transform is squared, the time-frequency matrix value of a signal part becomes larger, the time-frequency matrix value of a non-signal part becomes smaller, and particularly the value of the part is smaller than 1. The spectrogram is therefore more able to highlight the portions of the signal that are present with a higher frequency resolution than the short-time fourier transform. On the other hand, as can be seen from the analysis in section 2.2, although the spectrogram belongs to nonlinear time-frequency distribution, if there is no time-frequency collision between the signals, the overall spectrogram transform of the sum of the signals will not generate cross-term interference, which is a great advantage of the spectrogram. For WVD and SPWVD, if WVD is adopted to perform time-frequency analysis on the multi-frequency-hopping signal, the influence of a large number of cross item interferences is far greater than the influence of increased calculated amount and reduced time-frequency resolution brought by SPWVD, so that SPWVD is more suitable for the multi-frequency-hopping signal. Therefore, the combination of the spectrum and SPWVD is a better choice.
2.2 Combined time-frequency distribution improving method
The processing of the spectrogram in the existing SP & SPWVD combined time-frequency distribution is the direct square of the short-time Fourier transform. The short-time fourier transform cannot achieve both high time resolution and high frequency resolution due to the limitation of the window function, the short-time fourier transform using the long window has high frequency resolution but low time resolution, and the short-time fourier transform using the short window has high time resolution but low frequency resolution. This causes that it is difficult to find a suitable window length to obtain a compromise time-frequency resolution using a single window length spectrogram, so that it is often not the best result to directly obtain a spectrogram by squaring a short-time fourier transform.
Fig. 1 is a flowchart of a method for obtaining combined time-frequency distribution in multi-frequency-hopping signal scouting according to an embodiment of the present invention.
In this embodiment, as shown in fig. 1, the method for obtaining combined time-frequency distribution in multi-frequency-hopping signal reconnaissance of the present invention is characterized by comprising the following steps:
step S1: calculating an analysis signal z (t) of the multi-frequency-hopping signal y (t);
step S2: carrying out long-window short-time Fourier transform on the analytic signal z (t) to obtain a time-frequency transform matrix STFTh1(m, n), wherein m is a time point, n is a discrete frequency point, and the window function is a Hamming window;
step S3: carrying out short-window short-time Fourier transform on the analytic signal z (t) to obtain a time-frequency transform matrix STFTh2(m, n), the window function being a hamming window;
step S4: the time-frequency transformation matrix STFT obtained in the steps S2 and S3h1(m,n)、STFTh2(m, n) multiplying after modulus calculation to obtain a time frequency matrix SP (m, n) of the combined window spectrogram;
step S5: and (4) performing truncation processing on the time-frequency matrix SP (m, n) of the combined window spectrogram obtained in the step (S4) to obtain a time-frequency matrix SP' (m, n) of the combined window spectrogram:
wherein the truncation threshold epsilon is:
wherein eta is a threshold factor and can be adjusted according to the signal-to-noise ratio, M is the number of time sampling points, and N is the number of frequency points;
step S6: solving a smooth pseudo Wigner distribution time frequency matrix SPWVD (m, n) of the analytic signal z (t);
step S7: and performing Hadamard product on the time-frequency matrix SP' (m, n) of the truncated combined window spectrogram obtained in the step S5 and the SPWVD time-frequency matrix SPWVD (m, n) obtained in the step S5 to obtain combined time-frequency distribution (namely improved SP & SPWVD combined time-frequency distribution).
The invention discloses a method for acquiring combined time-frequency distribution in multi-hop signal reconnaissance, which is characterized in that Spectrogram processing is improved on the basis of the conventional SP & SPWVD, firstly, a Hamming window is selected as a window function to carry out long-window and short-window short-time Fourier transform, then, a long-window short-time Fourier transform with higher frequency resolution and a time-frequency transform matrix of a short-window short-time Fourier transform with higher time resolution are subjected to modulo multiplication to obtain a time-frequency matrix SP (m, n) of a combined window Spectrogram (SP, i.e. Spectrogram) and then, a time-frequency matrix of the combined window Spectrogram is subjected to truncation processing to obtain a time-frequency matrix SP '(m, n), and finally, the time-frequency matrix SP' (m, n) and the SPWVD are subjected to Hadamard product to obtain combined time-frequency distribution (namely, the improved SP & SPWVD combined time-frequency distribution). Compared with the existing SP & SPWVD combined time-frequency distribution, the time-frequency resolution is further improved, the time-frequency graph is clearer and more stable, and the calculated amount is not increased.
3. Simulation verification
Assuming that the receiver receives a complete frequency hopping signal within the observation time, and the background noise is white gaussian noise with a signal-to-noise ratio equal to 0dB, the simulation observation time is set to 0.4s and the sampling frequency is 4000Hz according to the mathematical model of the multi-frequency hopping signal in chapter ii, and two frequency hopping signals are generated: frequency hopping signal s1(t)Set to {1100, 1300, 1600, 1000, 1700, 1500, 1200, 1400}, in HZ, with a hop period Th1 of 0.05 s; frequency hopping signal s2The hopping frequency set of (t) is set to {100, 300, 600, 400, 200}, the unit is HZ, and the hopping period Th2 is 0.08 s.
3.1 time-frequency diagram comparison
Fig. 2 to 6 are time-frequency diagrams of different combinations of time-frequency distributions, in which:
the STFT and WVD combined time-frequency distribution in FIG. 2 has more cross-term interference and low time-frequency resolution;
the STFT & PWVD combined time-frequency distribution in fig. 3 suppresses a large amount of cross-term interference, however, the time-frequency resolution is not high, compared to the STFT & PWVD combined time-frequency distribution;
the STFT & SPWVD combined time-frequency distribution in FIG. 4 almost inhibits all cross term interference, but the time-frequency resolution is not high and is not improved;
the improvement effect of the existing SP & SPWVD combined time-frequency distribution in FIG. 5 is not obvious compared with that of the STFT & SPWVD time-frequency diagram;
fig. 6 is a combined time-frequency distribution diagram obtained by the present invention, i.e. an improved SP & SPWVD combined time-frequency distribution diagram, and as can be seen from fig. 6, compared with other combined time-frequency distributions, the time-frequency resolution is higher, the time-frequency diagram is clear and stable, and is suitable for time-frequency analysis of multi-frequency-hopping signals.
3.2 information entropy comparison
In section 3.1, qualitative comparison of time-frequency graphs of various combined time-frequency distributions is performed, that is, the quality of time-frequency analysis effect is judged from visual perception on the time-frequency graphs, and the method for obtaining the combined time-frequency distributions in multi-frequency-hopping signal reconnaissance is verified. The information entropy is applied to quantitatively measure the effect of time-frequency distribution of various combinations, and the performance of the invention is verified from the quantitative point of view.
The better the time-frequency aggregation of the time-frequency distribution, the denser the distribution of the energy in the time-frequency plane; under the same time-frequency aggregation property, the smaller the cross term of the time-frequency distribution is, the more concentrated the distribution of the energy of the time-frequency distribution in the time-frequency plane is. Therefore, whether it is time-frequency aggregation or the severity of cross terms, it can be described by the sparsity of the distribution on the time-frequency plane. And a very effective tool to describe this sparsity is entropy. The performance of the time-frequency distribution can be considered uniformly by utilizing the information entropy, so that the performance of the time-frequency distribution is described quantitatively. For the same signal, it is desirable that the information entropy of the time-frequency distribution is as small as possible. According to the algorithm of the information entropy, the information entropy of different combined time-frequency distributions is obtained as shown in table 1:
| combined time-frequency distribution | Entropy of information |
| STFT&WVD | 0.6774 |
| STFT&PWVD | 0.6752 |
| STFT&SPWVD | 0.6737 |
| SP&SPWVD | 0.6704 |
| The present invention, improved SP&SPWVD | 0.6652 |
TABLE 1
As can be seen from table 1, the information entropy values of the combined time-frequency distribution are relatively small, because the performance of the combined time-frequency distribution is improved relative to that of a single time-frequency distribution, and the information entropy value is not difficult to be obtained from the information entropy specific algorithm given in the literature, and is closely related to the number of signal sampling points, the comparison of the information entropy can only be performed for a specific signal. In the present invention, the results shown in table 1 are information entropy values of time-frequency distributions of different combinations of multiple frequency hopping signals under the simulation conditions set in the foregoing. Therefore, it can be seen that the information entropy of the improved SP & SPWVD combined time-frequency distribution is reduced compared with the general SP & SPWVD combined time-frequency distribution algorithm, which indicates that the performance of the invention is better, and the result is consistent with the comparison result of the time-frequency diagram in section 3.1.
5 summary of the invention
In the field of frequency hopping communication reconnaissance, along with the continuous improvement of human informatization degree, an electromagnetic environment is increasingly complex, and multiple frequency hopping signals generally exist in reality. A single time-frequency analysis method often cannot meet the time-frequency analysis requirements for multi-frequency-hopping signals, and thus various combined time-frequency distribution algorithms are applied. However, the SP & SPWVD combined time-frequency distribution algorithm is a method suitable for engineering implementation, but its advantages in time-frequency diagram effect are not obvious compared with other combined time-frequency distribution algorithms. The invention is improved based on SP and SPWVD combined time-frequency distribution, and obviously improves the algorithm performance without increasing the algorithm operation amount. Simulation results show that compared with other combined time-frequency distributions, the time-frequency diagram is clearer, more stable and finer, and has the lowest information entropy value from the viewpoint of information entropy.
Although illustrative embodiments of the present invention have been described above to facilitate the understanding of the present invention by those skilled in the art, it should be understood that the present invention is not limited to the scope of the embodiments, and various changes may be made apparent to those skilled in the art as long as they are within the spirit and scope of the present invention as defined and defined by the appended claims, and all matters of the invention which utilize the inventive concepts are protected.
Claims (1)
1. A method for obtaining combined time-frequency distribution in multi-frequency-hopping signal reconnaissance is characterized by comprising the following steps:
(1) determining an analysis signal z (t) of the multi-frequency-hopping signal y (t);
(2) and performing long-window short-time Fourier transform on the analytic signal z (t) to obtain a time-frequency transform matrix STFTh1(m, n), wherein m is a time point, n is a discrete frequency point, and the window function is a Hamming window;
(3) and carrying out short-window short-time Fourier transform on the analytic signal z (t) to obtain a time-frequency transform matrix STFTh2(m, n), the window function being a hamming window;
(4) transforming the time-frequency matrix STFTh1(m,n)、STFTh2(m, n) multiplying after modulus calculation to obtain a time frequency matrix SP (m, n) of the combined window spectrogram;
(5) and cutting off the time frequency matrix SP (m, n) of the combined window spectrogram to obtain a time frequency matrix SP' (m, n) of the combined window spectrogram:
wherein the truncation threshold epsilon is:
wherein eta is a threshold factor and can be adjusted according to the signal-to-noise ratio, M is the number of time sampling points, and N is the number of frequency points;
(6) solving a smooth pseudo Wigner distribution time-frequency matrix SPWVD (m, n) of the analytic signal z (t);
(7) and performing Hadamard product on the time frequency matrix SP' (m, n) of the combined window spectrogram after the truncation processing and the SPWVD time frequency matrix SPWVD (m, n) to obtain combined time frequency distribution, namely improved SP & SPWVD combined time frequency distribution.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910640977.2A CN110336587B (en) | 2019-07-16 | 2019-07-16 | Method for acquiring combined time-frequency distribution in multi-frequency-hopping signal reconnaissance |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910640977.2A CN110336587B (en) | 2019-07-16 | 2019-07-16 | Method for acquiring combined time-frequency distribution in multi-frequency-hopping signal reconnaissance |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110336587A CN110336587A (en) | 2019-10-15 |
| CN110336587B true CN110336587B (en) | 2021-10-22 |
Family
ID=68145135
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910640977.2A Active CN110336587B (en) | 2019-07-16 | 2019-07-16 | Method for acquiring combined time-frequency distribution in multi-frequency-hopping signal reconnaissance |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110336587B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111427070B (en) * | 2020-05-09 | 2023-03-14 | 电子科技大学 | GNSS anti-deception jamming method |
| CN111600630B (en) * | 2020-05-12 | 2021-07-06 | 中国电子科技集团公司第五十四研究所 | Frequency hopping signal detection method combining FFT (fast Fourier transform) with large and small points |
| JP2023549492A (en) * | 2020-11-11 | 2023-11-27 | ホアウェイ・テクノロジーズ・カンパニー・リミテッド | Signal transmission method, signal reception method, and device |
| CN112994741B (en) * | 2021-05-11 | 2021-07-23 | 成都天锐星通科技有限公司 | Frequency hopping signal parameter measuring method and device and electronic equipment |
| CN113489552B (en) * | 2021-05-21 | 2022-10-18 | 杭州电子科技大学 | Frequency hopping signal detection method based on time-frequency spectrum matrix local variance |
| CN117786398A (en) * | 2024-02-28 | 2024-03-29 | 杰创智能科技股份有限公司 | Frequency hopping signal characteristic identification method, system, electronic equipment and storage medium |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104485979A (en) * | 2014-12-09 | 2015-04-01 | 西安电子科技大学 | Blind estimation method for underdetermined hybrid frequency hopping parameters based on time frequency diagram correction |
| CN107885698A (en) * | 2017-10-23 | 2018-04-06 | 中国地质大学（武汉） | A kind of BGabor NSPWVD three-dimensional Time-Frequency Analysis Methods for noisy ENPEMF signals |
-
2019
- 2019-07-16 CN CN201910640977.2A patent/CN110336587B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104485979A (en) * | 2014-12-09 | 2015-04-01 | 西安电子科技大学 | Blind estimation method for underdetermined hybrid frequency hopping parameters based on time frequency diagram correction |
| CN107885698A (en) * | 2017-10-23 | 2018-04-06 | 中国地质大学（武汉） | A kind of BGabor NSPWVD three-dimensional Time-Frequency Analysis Methods for noisy ENPEMF signals |
Non-Patent Citations (2)
| Title |
|---|
| "组合图谱法在跳频信号时频分析中的应用";曾祥峰等;《计算机工程与应用》;20130701(第13期);第89-90页 * |
| "跳频信号的侦察技术研究";陈利虎;《信息科技辑》;20100415(第04期);正文第28-29、54-55页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110336587A (en) | 2019-10-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110336587B (en) | Method for acquiring combined time-frequency distribution in multi-frequency-hopping signal reconnaissance | |
| Shi et al. | A novel fractional wavelet transform and its applications | |
| CN108469560B (en) | Electromagnetic interference objective complexity evaluation method based on rapid S-transform time-frequency space model | |
| Rickett | Amplitude-modulated noise-an empirical model for the radio radiation received from pulsars | |
| Wang et al. | Selection of window length for singular spectrum analysis | |
| CN108594177A (en) | Based on radar signal modulation system analysis method, the signal processing system for improving HHT | |
| CN103163505A (en) | Time-varying narrow-band interference suppression method based on joint approximate diagonalization of eigen-matrices (JADE) | |
| Buch et al. | Towards real-time impulsive RFI mitigation for radio telescopes | |
| Divakaran et al. | Wavelet based spectrum sensing techniques for cognitive radio-a survey | |
| McDonald et al. | A statistical and physical mechanisms-based interference and noise model for array observations | |
| CN114488055A (en) | Coherent processing method suitable for carrier frequency and repetition frequency combined agile radar | |
| Wang et al. | An adaptive time-frequency filtering algorithm for multi-component LFM signals based on generalized S-transform | |
| Zhu et al. | Single channel source separation and parameter estimation of multi-component PRBCPM-SFM signal based on generalized period | |
| Zhang et al. | Optimization of adaptive MTI filter | |
| Kumari et al. | Estimation of intrapulse modulation parameters of LPI radar under noisy conditions | |
| Long et al. | Signal-to-noise ratio in stretch processing | |
| Kim et al. | Comparison of methods for parameter estimation of frequency hopping signals | |
| Gokalp et al. | Min-norm method for estimating delay and Doppler profiles from FMCW channel data with in-band interference | |
| Wong et al. | Improved target detection in spiky sea clutter using sparse signal separation | |
| Shen et al. | Radar Waveform Recognition Based on Time-frequency Images Under Deteriorating Channel Conditions | |
| Detka et al. | The transitory evolutionary spectrum | |
| Ni et al. | Detection of underwater carrier-free pulse based on time-frequency analysis | |
| Li et al. | Compressive frequency estimation for frequency hopping signal | |
| Flandrin et al. | Recent advances in theory and methods for nonstationary signal analysis | |
| Zhang et al. | Parameter Estimation of Multi-Frequency Hopping Signals Based on FastICA Algorithm |
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 |