CN111611686A - Detection method for communication signal time-frequency domain - Google Patents
Detection method for communication signal time-frequency domain Download PDFInfo
- Publication number
- CN111611686A CN111611686A CN202010291846.0A CN202010291846A CN111611686A CN 111611686 A CN111611686 A CN 111611686A CN 202010291846 A CN202010291846 A CN 202010291846A CN 111611686 A CN111611686 A CN 111611686A
- Authority
- CN
- China
- Prior art keywords
- time
- frequency
- signal
- image
- frequency domain
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000004891 communication Methods 0.000 title claims abstract description 19
- 238000001514 detection method Methods 0.000 title claims abstract description 17
- 238000000034 method Methods 0.000 claims abstract description 23
- 230000009466 transformation Effects 0.000 claims abstract description 16
- 238000012545 processing Methods 0.000 claims abstract description 12
- 238000004458 analytical method Methods 0.000 claims abstract description 6
- 239000011159 matrix material Substances 0.000 claims description 16
- 238000005070 sampling Methods 0.000 claims description 16
- 238000009826 distribution Methods 0.000 claims description 13
- 238000004364 calculation method Methods 0.000 claims description 5
- 230000009467 reduction Effects 0.000 claims description 4
- 230000008569 process Effects 0.000 abstract description 3
- 238000004088 simulation Methods 0.000 description 9
- 238000010586 diagram Methods 0.000 description 5
- 238000011426 transformation method Methods 0.000 description 4
- 238000006243 chemical reaction Methods 0.000 description 3
- 238000012216 screening Methods 0.000 description 2
- 108010076504 Protein Sorting Signals Proteins 0.000 description 1
- 238000005311 autocorrelation function Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 239000004576 sand Substances 0.000 description 1
- 230000003595 spectral effect Effects 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
- 230000001131 transforming effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R23/00—Arrangements for measuring frequencies; Arrangements for analysing frequency spectra
- G01R23/02—Arrangements for measuring frequency, e.g. pulse repetition rate; Arrangements for measuring period of current or voltage
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R23/00—Arrangements for measuring frequencies; Arrangements for analysing frequency spectra
- G01R23/16—Spectrum analysis; Fourier analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/10—Numerical modelling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/02—Preprocessing
- G06F2218/04—Denoising
Abstract
The invention discloses a detection method of a time-frequency domain of a communication signal, which utilizes a time-frequency analysis and image recognition method to carry out analysis processing, completes the detection of the initial time and the frequency of the time-frequency domain of the communication signal, and obtains the specific position of the signal. The invention processes the time-frequency domain image after the transformation by methods such as image recognition and the like on the basis of the time-frequency domain transformation of the signal, and then calculates the initial position of the time-frequency domain of the obtained matched image by an algorithm, thereby obtaining the initial positions of all time domains and frequency domains in the whole segment of signal.
Description
Technical Field
The invention relates to time-frequency analysis and image processing technology, in particular to a detection method of a communication signal time-frequency domain, aiming at improving the efficiency of signal detection.
Background
The method is related to signal-to-noise ratio and threshold value, has certain error, and the energy detection can only be carried out in a single direction, namely frequency domain or time domain, and needs to carry out detection twice in two dimensions.
The Fourier transform is commonly used for the conversion between a time domain and a frequency domain, the duration and the existing position of signals corresponding to different time of frequency domain components cannot be seen through the traditional Fourier transform, the frequency range can be obtained only through an energy detection method, the time-frequency transform of the signals provides the joint distribution of the time domain and the frequency domain of the signals, and the change process of the signals along with the time can be well reflected.
Image recognition processing has been rapidly developed in recent years, and is widely used in various fields. The image recognition is based on image features, the image features comprise color features, texture features, shape features, local feature points and the like, the color features are the most basic common features and comprise histograms, color sets, color moments and the like, the local features have good stability and are not easily interfered by external environments, and the method is an effective mode for simplifying image data.
Disclosure of Invention
In view of the above-mentioned drawbacks of the prior art, an object of the present invention is to provide a method for detecting a time-frequency domain of a communication signal.
In order to achieve the purpose, the invention adopts the following technical scheme:
a detection method of communication signal time-frequency domain, utilize time-frequency analysis and image recognition method to analyze and process, finish the initial time of the time-frequency domain of the communication signal and detection of the frequency of place, receive the concrete position where the signal is located; the method comprises the following specific steps:
1-1, receiving a signal, carrying out noise reduction processing on the signal, and then sampling the signal with a sampling rate fs；
1-2, performing time-frequency transformation of different methods on the denoised signals, setting a threshold value of the time-frequency transformation, outputting a calculated time-frequency image and a calculated time-frequency point image, and storing information of the time-frequency image and the time-frequency point matrix;
1-3, calculating the obtained time-frequency point matrix, and outputting frequency information within the signal duration time of the time-frequency point matrix to obtain all time and frequency starting point information;
1-4, graying or binarizing the obtained time-frequency image, storing the required part, calculating and processing the continuously existing signals, storing, and finally outputting the time and frequency of the signal.
Because the communication signals have relatively fixed frequency bands and time gaps, different time-frequency domain transformation methods can set parameters suitable for the signals according to the characteristics of the communication signals, so that the time-frequency diagram with higher quality is obtained while system computing resources are saved, and the subsequent image processing is easier to realize. In the step 1-2, two methods of short-time Fourier transform and Wigner-Ville distribution are used for synchronously performing time-frequency transform, each method obtains two different results through respective calculation, and the results are processed and then integrated, so that the results are more accurate.
Compared with the prior art, the invention has the following advantages:
the invention utilizes the time-frequency transformation and the image identification method to carry out the detection function of the communication signal, utilizes the time-frequency transformation method to enable the time domain and the frequency domain of the synchronous detection signal to become possible, and is a more efficient and accurate signal detection method.
Drawings
Fig. 1 is a flow chart of a communication signal time-frequency domain detection method according to the invention.
FIG. 2 is an original time domain image of a simulated signal with a sampling rate of 2KHz, a time domain image after noise addition and low pass filtering.
FIG. 3 is a time-frequency domain diagram of a simulation signal with a sampling rate of 2KHz after Wigner-Ville distribution transformation.
FIG. 4 shows the required part of the simulated signal after cutting, with a sampling rate of 2 KHz.
FIG. 5 is a data point diagram of a simulated signal at 2KHz sampling rate after a short time Fourier transform.
FIG. 6 is an original time domain image of a simulated signal at a sample rate of 2MHz, a noise-and low-pass filtered time domain image.
FIG. 7 is a time-frequency domain diagram of a simulation signal with a sampling rate of 2MHz after Wigner-Ville distribution transformation.
FIG. 8 is a graph of time versus frequency for a simulated signal having a sample rate of 2 KHz.
FIG. 9 is a graph of time versus frequency for a simulated signal at a 2KHz sampling rate after a short time Fourier transform.
FIG. 10 is a graph of the time and frequency position of a simulated signal at 2KHz after Wigner-Ville distribution transformation.
Detailed Description
The following detailed description of the embodiments of the present invention is provided with reference to the accompanying drawings, but the present invention is not limited to the embodiments, and all similar structures and similar variations of the present invention can be adopted within the scope of the present invention.
The communication signal typically has a relatively fixed frequency range and duration with a minimum frequency f1Maximum frequency of f2With a duration of generally tsThe signal-to-noise ratio ranges from about-5 dB to about 30 dB.
As shown in fig. 1, a method for detecting a time-frequency domain of a communication signal includes the following steps:
1-1, receiving a signal, carrying out noise reduction processing on the signal, and then sampling the signal with a sampling rate fs；
1-2, performing time-frequency transformation of different methods on the denoised signals, setting a threshold value of the time-frequency transformation, outputting a calculated time-frequency image and a calculated time-frequency point image, and storing information of the time-frequency image and the time-frequency point matrix;
1-3, calculating the obtained time-frequency point matrix, and outputting frequency information within the signal duration time of the time-frequency point matrix to obtain all time and frequency starting point information;
1-4, graying or binarizing the obtained time-frequency image, storing the required part, calculating and processing the continuously existing signals, storing, and finally outputting the time and frequency of the signal.
After obtaining the signal, firstly, the signal is denoisedAt the moment, the noise is mostly thermal noise, the instantaneous value of the thermal noise follows Gaussian distribution, meanwhile, the power spectral density is uniform and is Gaussian white noise, and a signal with higher quality is obtained after the noise is reduced through a low-pass filter. The filtered signal is then sampled at a sampling rate fsConverting the time-frequency domain image into a digital signal to perform time-frequency domain transformation, wherein the quality of a time-frequency image depends on factors such as a time-frequency domain transformation method, window function setting and the like, and a better time-frequency image is obtained by changing the setting of the function and the parameter and is stored; and then preprocessing the stored image, binarizing the image to obtain a binarized image, reserving a required part for calculation processing, and storing the time and frequency containing part in an excel table according to a calculation result.
The signal in step 1-1 of the embodiment of the invention is obtained by MATLAB simulation, and the maximum value f of the frequency of the signal used by the simulation1Is 400Hz, minimum value f2100Hz, a duration of 10s, a single signal duration of 0.5s, a sampling frequency fsAt 2KHz as shown in figure 8. The signals are explained hereafter, and the rest signals can be obtained by parameter-adjustable simulation, as shown in fig. 6, and the time-frequency conversion results are shown in fig. 7. Minimum value f of signal frequency used for simulation at this time1Is 100KHz, maximum value f2200KHz, duration 0.01 s. The simulation signal is a bpsk signal, and the functional formula of the carrier signal is as follows:
wherein ω iscIs the carrier frequency and is,
is a phase controlled by a digital signal. Will signal Sk(t) and normalized bipolar digital signal sequence χkMultiplying to obtain the required simulation signal S1(t) a functional formula of:
when the baseband digital signal is 0, χkIs +1, at this time
Is 0; and when the baseband digital signal is 1, χkIs-1, at this time
Is pi; then S1(t) adding Gaussian white noise, wherein the noise is added through an MATLAB built-in function awgn, the signal-to-noise ratio is set to be 10dB, and the used signals are the signal-to-noise ratio to obtain a simulation signal S of a received signal2(t) of (d). Before time-frequency transformation, noise reduction is carried out, and because a communication signal has a specific range, the cut-off frequency is set to be half of the maximum sampling frequency when a low-pass filter is designed, namely the cut-off frequency is set
The design of the low-pass filter is not described herein, and as shown in fig. 2, the signal s (t) is obtained.
The two time-frequency transformation methods in the step 1-2 of the embodiment of the invention are short-time Fourier transform (STFT) and Wigner-Ville distribution. Wherein, the short-time Fourier transform represents the signal characteristic at a certain moment through a section of signal in a time window, the length of the window determines the time resolution and the frequency resolution of the spectrogram, the longer the window is, the longer the intercepted signal is, the longer the signal is, the higher the frequency resolution is after the Fourier transform is, and the worse the time resolution is; conversely, the shorter the window length, the shorter the intercepted signal, the poorer the frequency resolution, the better the time resolution, and the discretized short-time fourier expression:
where s (k) is the source signal, g (t) is a window function where the window length is chosen to be 2048, the number of overlapping points is 1920, and the sampling frequency is fsThe FFT point number is 2048 points, and the minimum threshold value is set by transforming the Blackman Harris window:
T＝-floor(log10(fs/1000)+1)*10
the unit is dB, where the floor () function represents rounding to minus infinity, as shown in fig. 5. Obtaining a data point diagram after conversion, storing the data points into an intermediate variable axis, wherein the intermediate variable axis is a vector of two columns, the first column stores the time of the signal which meets the threshold, the second column stores the frequency of the signal which meets the threshold, the communication signal has a fixed length, 0.5s is taken as an example, the signal is subdivided by 1s, and the frequency error is set to be 1 percent fsAnd storing the data into a new matrix point, obtaining the frequency starting data with the unit of 1s, and storing the data into a file signal position (data), as shown in fig. 9, obtaining the time and the frequency position of the signal, and comparing fig. 8 to see whether the result is accurate.
The Wigner-Ville distribution is quadratic transformation, window operation is not added in calculation, the mutual constraint between time domain resolution and frequency domain resolution is avoided, but the interference of cross terms exists, and the expression of WVD is as follows:
wherein
Is the instantaneous autocorrelation function R (t, τ) of the signal s (t). When the signal S (t) is S1(t)+S2(t), the corresponding Wigner-Ville distribution can be expressed as:
Ws(t,f)＝Ws1(t,f)+Ws2(t,f)+Ws1s2(t,f)+Ws2s1(t,f)
wherein:
these two terms become mutual Wigner-Ville distributions, which are complex valued and exist simultaneously:
thus Ws1s2(t,f)+Ws2s1(t, f) is real-valued, and the above formula can be abbreviated as:
Ws(t,f)＝Ws1(t,f)+Ws2(t,f)+2Re[Ws2s1(t,f)]
wherein the additional term 2Re [ W ]s2s1(t,f)]Generally referred to as cross terms, to solve the problem of cross terms, pseudo-smooth WVD is used here, which is a method of windowing and smoothing WVD, where a hamming window of 601 sample points is selected to window the distribution in time, a rectangular window of 305 points is selected to window the distribution in frequency, 600 frequency points are used for display, the threshold value is the same as that used for short-time fourier transform, and the energy level is set to a color image of 256 levels of colors, as shown in fig. 3. The image is saved as shown in fig. 4 and the data required by the image. And then screening energy points which meet the conditions according to threshold conditions, storing and calculating the time and frequency points of the energy points, wherein the energy points which meet the conditions are concentrated but have errors, because the signal duration is 0.5s and the frequency is 2000Hz, in order to reduce the errors, firstly, rounding the reserved decimal point one bit in time and accurately obtaining ten bits in frequency, screening all data which meet the conditions and respectively take 0.1s and 10Hz as one grade, and storing the data into an intermediate variable matrix. Finally, in order to judge whether the data are continuous signals at time intervals of 0.5s, a new matrix a is created, all the time data in the matrix are added with 0.1s, the frequency is kept unchanged, whether each row in the new matrix a exists in the original matrix is compared line by line, if the row does not exist, the row in the matrix is returned, the row is the time end point of the single signal and the frequency of the corresponding signal, then the data are stored in a result variable result, and the data are stored in a file signal position (image) through a cvswitch function, as shown in fig. 10.
Claims (2)
1. A detection method of a communication signal time-frequency domain is characterized in that a time-frequency analysis and image recognition method is used for carrying out analysis processing, the detection of the initial time and the frequency of the communication signal time-frequency domain is completed, and the specific position of the signal is obtained; the method comprises the following specific steps:
1-1, receiving a signal, carrying out noise reduction processing on the signal, and then sampling the signal with a sampling rate fs；
1-2, performing time-frequency transformation of different methods on the denoised signals, setting a threshold value of the time-frequency transformation, outputting a calculated time-frequency image and a calculated time-frequency point image, and storing information of the time-frequency image and the time-frequency point matrix;
1-3, calculating the obtained time-frequency point matrix, and outputting frequency information within the signal duration time of the time-frequency point matrix to obtain all time and frequency starting point information;
1-4, graying or binarizing the obtained time-frequency image, storing the required part, calculating and processing the continuously existing signals, storing, and finally outputting the time and frequency of the signal.
2. The method for detecting the time-frequency domain of the communication signal as claimed in claim 1, wherein in the step 1-2, the time-frequency transformation is performed synchronously by using two methods of short-time fourier transform and Wigner-Ville distribution, each method obtains two different results through respective calculation, and the results are processed and integrated respectively, so that the results are more accurate.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010291846.0A CN111611686A (en) | 2020-04-14 | 2020-04-14 | Detection method for communication signal time-frequency domain |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010291846.0A CN111611686A (en) | 2020-04-14 | 2020-04-14 | Detection method for communication signal time-frequency domain |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111611686A true CN111611686A (en) | 2020-09-01 |
Family
ID=72205625
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010291846.0A Pending CN111611686A (en) | 2020-04-14 | 2020-04-14 | Detection method for communication signal time-frequency domain |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111611686A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114553639A (en) * | 2022-02-21 | 2022-05-27 | 中国人民解放军国防科技大学 | Method for detecting and identifying Morse signal |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102693546A (en) * | 2011-03-23 | 2012-09-26 | 苏州大学 | Signal time frequency characteristic image generation method and device thereof |
| KR20180012487A (en) * | 2016-07-27 | 2018-02-06 | 국방과학연구소 | Method for estimating variables of missile target |
| 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 |
| CN110336638A (en) * | 2019-04-17 | 2019-10-15 | 四川大学 | A kind of short-term burst signal detecting method based on time-frequency figure |
-
2020
- 2020-04-14 CN CN202010291846.0A patent/CN111611686A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102693546A (en) * | 2011-03-23 | 2012-09-26 | 苏州大学 | Signal time frequency characteristic image generation method and device thereof |
| KR20180012487A (en) * | 2016-07-27 | 2018-02-06 | 국방과학연구소 | Method for estimating variables of missile target |
| 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 |
| CN110336638A (en) * | 2019-04-17 | 2019-10-15 | 四川大学 | A kind of short-term burst signal detecting method based on time-frequency figure |
Non-Patent Citations (4)
| Title |
|---|
| 于晓婉: ""基于广义S变换的通信信号分析方法研究"", 《中国优秀硕士学位论文全文数据库（信息科技辑）》 * |
| 易伟: ""基于时频分析的跳频信号检测"", 《中国优秀硕士学位论文全文数据库（信息科技辑）》 * |
| 来丽晗: ""复杂环境下跳频信号参数盲估计算法研究"", 《中国优秀硕士学位论文全文数据库（信息科技辑）》 * |
| 殷婕 等: ""基于图像处理的跳频信号参数估计"", 《通信与网络》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114553639A (en) * | 2022-02-21 | 2022-05-27 | 中国人民解放军国防科技大学 | Method for detecting and identifying Morse signal |
| CN114553639B (en) * | 2022-02-21 | 2024-02-27 | 中国人民解放军国防科技大学 | Morse signal detection and identification method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101919695B (en) | Electrocardiosignal QRS complex detection method based on wavelet transform | |
| CN112446323B (en) | HHT harmonic analysis method based on improved EMD modal aliasing and endpoint effect | |
| CN103454495A (en) | Self-adaptive high-precision fast spectral analysis method | |
| CN108427031B (en) | Inter-harmonic detection method based on polynomial fitting and non-interference region division | |
| CN110068727B (en) | Single-frequency signal frequency estimation method based on Candan-Rife comprehensive interpolation | |
| CN114025379B (en) | Broadband multi-signal detection method, device and equipment | |
| CN103346845A (en) | Fast Fourier transform-based blind frequency spectrum sensing method and apparatus | |
| CN109379310B (en) | Rife-Quinn synthesis-based MPSK signal carrier frequency estimation method | |
| CN111934711A (en) | Parameter estimation method of time-frequency aliasing frequency hopping signal | |
| CN111562438B (en) | Sinusoidal signal frequency estimation method and device based on FFT and phase difference | |
| CN105429719A (en) | Strong interference signal detection method based on power spectrum and multiple dimensioned wavelet transformation analysis | |
| CN113702703B (en) | Weak signal detection and identification method and system | |
| CN111611686A (en) | Detection method for communication signal time-frequency domain | |
| CN113447893B (en) | Radar pulse signal frequency spectrum automatic detection method, system and medium | |
| CN110632563B (en) | Intra-pulse frequency coding signal parameter measuring method based on short-time Fourier transform | |
| Djurovic | Estimation of the sinusoidal signal frequency based on the marginal median DFT | |
| CN106777505A (en) | The frequency estimating methods and device of the robust of the undersampled signal based on frequency deviation identification | |
| CN109117698B (en) | Noise background estimation method based on minimum mean square error criterion | |
| CN113627398B (en) | Signal characteristic detection method based on adaptive reconstruction filtering | |
| CN112883787B (en) | Short sample low-frequency sinusoidal signal parameter estimation method based on spectrum matching | |
| CN113848391A (en) | Pulse signal detection and extraction method | |
| CN113238200A (en) | Radar chirp signal classification method based on validity verification | |
| CN107576842B (en) | Broadband synchronous sampling method | |
| CN113050131A (en) | Capturing method based on preprocessing FFT and barrier effect correction | |
| CN111551785A (en) | Frequency and harmonic detection method based on unscented Kalman filter |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200901 |