CN110336638A - A kind of short-term burst signal detecting method based on time-frequency figure - Google Patents
A kind of short-term burst signal detecting method based on time-frequency figure Download PDFInfo
- Publication number
- CN110336638A CN110336638A CN201910310493.1A CN201910310493A CN110336638A CN 110336638 A CN110336638 A CN 110336638A CN 201910310493 A CN201910310493 A CN 201910310493A CN 110336638 A CN110336638 A CN 110336638A
- Authority
- CN
- China
- Prior art keywords
- signal
- time
- frequency
- detection
- short
- 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
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L1/00—Arrangements for detecting or preventing errors in the information received
- H04L1/0001—Systems modifying transmission characteristics according to link quality, e.g. power backoff
- H04L1/0036—Systems modifying transmission characteristics according to link quality, e.g. power backoff arrangements specific to the receiver
Landscapes
- Engineering & Computer Science (AREA)
- Quality & Reliability (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Testing, Inspecting, Measuring Of Stereoscopic Televisions And Televisions (AREA)
Abstract
The invention discloses a kind of short-term burst signal detecting methods based on time-frequency figure, belong to signal detection field.The present invention overcomes because of burst signal time-division and frequency division multiplexing caused by signal missing inspection problem and calculate the excessively high problem of cost.The technical solution of the present invention is as follows: 1) draw the time-frequency figure of input signal；2) input signal all sizes being likely to occur in time-frequency figure are calculated according to the priori knowledge of input signal, according to these corresponding rectangle frames of size design size；3) it using the signal in these frames detection time-frequency figure, obtains the initial time of each signal and terminates the time.In contrast to identical signal data, compared with original signal energy detection method, the signal more 70% that the method that the present invention uses detects calculates cost and composes lower than higher order cumulants, reduces signal omission factor and calculates cost.
Description
Technical field
The invention belongs to signal detection technique fields, are a kind of methods for carrying out signal detection using signal time-frequency figure, main
If being used for the detection of short-term burst signal, figure specifically is carried out using the color difference that the signal in time-frequency figure and noise occur
As identification is to realize signal detection.
Background technique
Burst communication is different from common communications, is randomly carried out data transmission using burst mode, and because burst is logical
The reason that the signal length of letter is shorter, sending device mostly minimizes, signal-to-noise ratio is also generally lower than 12dB, and short-term burst is believed
Number there are time-division and frequency division multiplexing, thus receiving end can not obtain burst signal initial time and other signal parameters.
After receiving end must accurately be detected and capture burst signal, the identification of parameter Estimation and modulating mode just can be carried out, finally
Burst signal is demodulated, the information of transmitting terminal transmission is obtained.
Currently, short-term burst signal detecting method mainly includes method based on linear theory and based on nonlinear theory
Method.Method based on linear theory mainly have traditional energy measuring method, frequency domain detection method and it is proposed in recent years when frequency division
Analysis method.Energy measuring method this theory will carry out signal detection, the party greatly using the pure noise energy of energy ratio in the presence of signal
Method algorithm is simple, easy to accomplish, and does not need any prior information, but very sensitive to the changed power of signal and noise,
The consequence for be easy to causeing verification and measurement ratio low in the environment of low signal-to-noise ratio.Frequency domain detection method is believed according to the period is extracted from ambient noise
Number characteristic frequency detect signal, can be detected under low signal-to-noise ratio environment, but for short-term burst signal detection because
In the presence of time division multiplexing and frequency division multiplexing, signal only cannot be distinguished using one latitude of frequency, therefore short-term burst can not be believed
Number detection effect is bad.Time-frequency Analysis by time signal by specific transformed mappings to a two-dimentional time-frequency plane, when
In frequency domain the T/F union feature of signal comprehensively can be understood in the property of time domain and frequency domain by observation signal simultaneously, when
Common time frequency analyzing tool has Short Time Fourier Transform and wavelet transformation in frequency analysis, finds out short-term burst using time frequency analysis
The highest point of the time-frequency spectrum of signal and centered on highest point toward each extension signal bandwidth in two sides half, for short-term burst believe
For number, two adjacent signal distances on frequency domain are close, adopt this method and are easy because caused by the fluctuation of noise
Two bars are mistaken for a bars by burr.
Main nonlinear theory signal detecting method mainly has high order equilibrium, neural network, empirical mode decomposition, mixes
Ignorant theoretical method.Linear theory signal detecting method loses caused by signal when extracting signal and crossing noise filtering, and non-thread
Property theory can avoid this influence well, nonlinear theory is suitable for detecting signal unstable, under nonequilibrium condition, but
Neural network need a large amount of training sample, empirical mode decomposition, high order equilibrium, chaology method require to signal into
Row high-order calculates, this will increase the calculating cost of signal, is unfavorable for the real-time of signal detection.
Summary of the invention
The problem of for above-mentioned existing signal detecting method, the present invention provide a kind of the low signal-to-noise ratio the case where
Lower detection effect is lower than the detection method of nonlinear theory signal detecting method better than conventional energy method, computation complexity, solves
Traditional linear theory method verification and measurement ratio is lower under low signal-to-noise ratio, and the problem that nonlinear method computation complexity is high.
The technical solution adopted by the present invention to solve the technical problems is: 1) drawing the time-frequency figure of input signal；2) basis
The priori knowledge of input signal calculates input signal all sizes being likely to occur in time-frequency figure, big according to these size designs
Small corresponding rectangle frame；3) using the signal in these frames detection time-frequency figure, when obtaining the initial time and termination of each signal
Between.
The present invention detects 70% signal more under conditions of identical signal than signal energy detection method, and calculates
Time cost be lower than Higher-Order Cumulants, therefore in verification and measurement ratio and to be calculated as present aspect advantageous.
Detailed description of the invention
The present invention will be further described with reference to the accompanying drawing.
Fig. 1 is detection method schematic diagram of the invention.
Fig. 2 is the flow chart of picture input signal time-frequency figure.
Fig. 3 is picture of the time-frequency figure of a wherein segment signal after binary conversion treatment.
Fig. 4 is that wherein white area is signal to the testing result based on time-frequency figure signal detecting method.
Specific embodiment
The present invention will be further described With reference to embodiment, with certain signal duration (ms) for x1、x2, symbol
Rate (symbol/s): for y1、y2Burst signal illustrate.
1. drawing the time-frequency figure of input signal
Firstly, input signal data is carried out segment processing, every segment data length is 150ms, x1、x2Respectively less than 150ms makes picture
The signal block size in time-frequency figure out is suitable, not only verification and measurement ratio will not be caused low because of block is too small, but also will not be because of too big
And cause detection time long；Then, Short Time Fourier Transform, the Short Time Fourier Transform of signal are done to every segment signal are as follows:
(1)
Wherein x (n) is the sampled data of signal,For window function, the time-frequency figure of signal is obtained by calculation,
The middle time is abscissa, and frequency is ordinate；Then, in order to reduce influence of the noise to signal detecting result, to signal when
Frequency figure does the processing of amplitude nonlinear change, the i.e. big part of amplification amplitude, reduces the small part of amplitude；Finally, in order to calculate letter
Just, the amplitude in time-frequency figure is subjected to binary conversion treatment, i.e., is considered noise when amplitude is 0, amplitude is considered signal when being 1.
2. designing rectangle frame according to the priori knowledge of input signal
Because of the feature of burst signal are as follows: signal duration (ms): x1、x2；Character rate (symbol/s): y1、y2.Due to signal
Character rate it is corresponding with bandwidth, therefore signal is the rectangle of following size: x in time-frequency figure1*y1、x1*y2、x2*y1、x2*y2。
The size and burst signal of rectangle frame may occupy in the same size in time-frequency figure.
3. carrying out signal detection using rectangle frame
It is sequentially overlapped movement on time-frequency figure with 4 kinds of rectangle frames being calculated respectively, calculates the pixel that amplitude is 1 in rectangle frame
Point.When the pixel number that amplitude is 1 is more than certain threshold value, it is believed that have signal appearance in the rectangle frame, record the rectangle
The position of frame.At this point, the abscissa of the leftmost side of rectangle frame corresponds to the initial time of the signal, the abscissa of the rightmost side is corresponding should
The end time of signal.
4. the result of output signal detection
According to the rectangle frame position for detecting signal of record, the biggish several identical rulers of percentage are overlapped firstly, for position
Very little rectangle frame, the largest number of rectangle frames of pixel that amplitude is 1 in choice box are deleted as signal, and by remaining rectangle frame
It removes；Secondly, the case where being included in large scale rectangle frame for small size rectangular frame, selects large scale rectangle frame for signal, delete
All small size rectangular frames being included；Finally, exporting the abscissa at left and right sides of remaining rectangle frame, left side abscissa is letter
Number initial time, right side abscissa be signal end time.
Claims (4)
1. a kind of short-term burst signal detecting method based on time-frequency figure, comprises the concrete steps that:
1) input signal is carried out segment processing, is then Fourier in short-term to every segment signal by the time-frequency figure for drawing input signal
Transformation, obtains the time-frequency figure of signal, then does the processing of amplitude nonlinear change to the time-frequency figure of signal, by the amplitude in time-frequency figure into
Row binary conversion treatment;
2) rectangle frame is designed according to the priori knowledge of input signal, it is corresponding by all possible duration of input signal and bandwidth Design
The rectangle frame of size;
3) signal detection is carried out using rectangle frame, is sequentially overlapped movement on time-frequency figure with the rectangle frame being calculated, works as amplitude
When more than certain threshold value, it is believed that have signal appearance in the rectangle frame;
4) output signal detection as a result, output rectangle frame at left and right sides of abscissa, left side abscissa be signal starting when
Between, right side abscissa is the end time of signal;
It is characterized by: signal detection is converted into image recognition, signal detection is carried out using the means of image recognition, compared to
Traditional time-frequency domain detection algorithm, reduces detection complexity, improves the verification and measurement ratio of signal.
2. the short-term burst signal detecting method according to claim 1 based on time-frequency figure, it is characterised in that: in the first step
Using short time discrete Fourier transform, signal detection is indicated with image, carries out signal inspection in conjunction with image-recognizing method to reach
The purpose of survey.
3. the short-term burst signal detecting method according to claim 1 based on time-frequency figure, it is characterised in that: second step benefit
Rectangle frame design is carried out with the priori knowledge of burst signal, all blocks appeared in time-frequency figure can be met with size,
This avoid be a plurality of signal for a bars erroneous detection when detection or be a bars by a plurality of signal erroneous detection.
4. the short-term burst signal detecting method according to claim 1 based on time-frequency figure, it is characterised in that: in the 4th step
The transverse and longitudinal coordinate of time-frequency figure is converted into time and frequency, to realize signal beginning and ending time and Frequency Estimation, reaches signal inspection
The purpose of survey.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910310493.1A CN110336638A (en) | 2019-04-17 | 2019-04-17 | A kind of short-term burst signal detecting method based on time-frequency figure |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910310493.1A CN110336638A (en) | 2019-04-17 | 2019-04-17 | A kind of short-term burst signal detecting method based on time-frequency figure |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110336638A true CN110336638A (en) | 2019-10-15 |
Family
ID=68139242
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910310493.1A Pending CN110336638A (en) | 2019-04-17 | 2019-04-17 | A kind of short-term burst signal detecting method based on time-frequency figure |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110336638A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111611686A (en) * | 2020-04-14 | 2020-09-01 | 上海大学 | Detection method for communication signal time-frequency domain |
| CN114679357A (en) * | 2022-02-24 | 2022-06-28 | 瑞玛思特(深圳)科技有限公司 | Method for detecting signal Burst |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080119966A1 (en) * | 2000-09-08 | 2008-05-22 | Breed David S | Wireless Sensing and Communication System for Traffic Lanes |
| CN104007318A (en) * | 2014-06-17 | 2014-08-27 | 中国科学院电子学研究所 | Signal time-frequency function obtaining method |
| CN105072067A (en) * | 2015-08-27 | 2015-11-18 | 西安电子科技大学 | Method for detecting burst-mode signal in low signal to noise ratio |
| CN105703850A (en) * | 2016-03-24 | 2016-06-22 | 电子科技大学 | Short-time-fourier-transform-based edge detection method for data chain signal |
| CN107945817A (en) * | 2017-11-15 | 2018-04-20 | 广东顺德西安交通大学研究院 | Heart and lung sounds signal sorting technique, detection method, device, medium and computer equipment |
| CN109286457A (en) * | 2018-09-04 | 2019-01-29 | 国网山东省电力公司经济技术研究院 | TWACS uplink signal detection method based on wavelet analysis |
| CN109379311A (en) * | 2018-09-30 | 2019-02-22 | 中国人民解放军战略支援部队信息工程大学 | Ultrashort wave signal specific recognition methods based on convolutional neural networks |
-
2019
- 2019-04-17 CN CN201910310493.1A patent/CN110336638A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080119966A1 (en) * | 2000-09-08 | 2008-05-22 | Breed David S | Wireless Sensing and Communication System for Traffic Lanes |
| CN104007318A (en) * | 2014-06-17 | 2014-08-27 | 中国科学院电子学研究所 | Signal time-frequency function obtaining method |
| CN105072067A (en) * | 2015-08-27 | 2015-11-18 | 西安电子科技大学 | Method for detecting burst-mode signal in low signal to noise ratio |
| CN105703850A (en) * | 2016-03-24 | 2016-06-22 | 电子科技大学 | Short-time-fourier-transform-based edge detection method for data chain signal |
| CN107945817A (en) * | 2017-11-15 | 2018-04-20 | 广东顺德西安交通大学研究院 | Heart and lung sounds signal sorting technique, detection method, device, medium and computer equipment |
| CN109286457A (en) * | 2018-09-04 | 2019-01-29 | 国网山东省电力公司经济技术研究院 | TWACS uplink signal detection method based on wavelet analysis |
| CN109379311A (en) * | 2018-09-30 | 2019-02-22 | 中国人民解放军战略支援部队信息工程大学 | Ultrashort wave signal specific recognition methods based on convolutional neural networks |
Non-Patent Citations (1)
| Title |
|---|
| 殷婕: ""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 |
|---|---|---|---|---|
| CN111611686A (en) * | 2020-04-14 | 2020-09-01 | 上海大学 | Detection method for communication signal time-frequency domain |
| CN114679357A (en) * | 2022-02-24 | 2022-06-28 | 瑞玛思特(深圳)科技有限公司 | Method for detecting signal Burst |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107968689B (en) | Perception identification method and device based on wireless communication signals | |
| CN108764077B (en) | Digital signal modulation classification method based on convolutional neural network | |
| Wang et al. | Radar emitter recognition based on the short time Fourier transform and convolutional neural networks | |
| CN110099019B (en) | LoRa modulation signal detection method based on deep learning | |
| CN108737318B (en) | OFDM signal identification method and system based on signal structure characteristics | |
| CN110336638A (en) | A kind of short-term burst signal detecting method based on time-frequency figure | |
| CN105354535B (en) | A kind of special Low Voltage Differential Signal Image Acquisition and processing method | |
| CN103607370B (en) | Credibility evaluation method for complex BPSK signal blind processing result | |
| CN112784835B (en) | Method and device for identifying authenticity of circular seal, electronic equipment and storage medium | |
| CN105785324A (en) | MGCSTFT-based chirp signal parameter estimation method | |
| CN105160342A (en) | HMM-GMM-based automatic word picture splitting method and system | |
| CN104869096B (en) | Bootstrap-based BPSK signal blind processing result credibility test method | |
| CN113780106A (en) | Deep learning signal detection method based on radio waveform data input | |
| CN114520736A (en) | Internet of things security detection method, device, equipment and storage medium | |
| CN115620316A (en) | Image recognition method, device, equipment and medium based on wireless federal learning | |
| CN102547477B (en) | Video fingerprint method based on contourlet transformation model | |
| CN115861210A (en) | Transformer substation equipment abnormity detection method and system based on twin network | |
| CN109190605B (en) | Human body continuous action counting method based on CSI | |
| CN105389820A (en) | Infrared image definition evaluating method based on cepstrum | |
| US9996767B2 (en) | Methods for automated chart analysis | |
| Panetta et al. | A human visual “no-reference” image quality measure | |
| US20090248336A1 (en) | Analyzer for signal anomalies | |
| Wang et al. | A spatiotemporal multi-stream learning framework based on attention mechanism for automatic modulation recognition | |
| CN104392252B (en) | Emitter Recognition and device | |
| CN102546116B (en) | Based on the Frequency Hopping Signal blind checking method of compressed sensing |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20191015 |