CN116992352B - Method for generating electromagnetic signal classification recognition data set, recognition method and device - Google Patents
Method for generating electromagnetic signal classification recognition data set, recognition method and device Download PDFInfo
- Publication number
- CN116992352B CN116992352B CN202311253730.8A CN202311253730A CN116992352B CN 116992352 B CN116992352 B CN 116992352B CN 202311253730 A CN202311253730 A CN 202311253730A CN 116992352 B CN116992352 B CN 116992352B
- Authority
- CN
- China
- Prior art keywords
- electromagnetic signal
- pulse
- dataset
- generating
- signal classification
- 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
- 238000000034 method Methods 0.000 title claims abstract description 44
- 238000005070 sampling Methods 0.000 claims abstract description 26
- 238000006243 chemical reaction Methods 0.000 claims abstract description 18
- 230000005855 radiation Effects 0.000 claims description 23
- 230000000630 rising effect Effects 0.000 claims description 14
- 238000009499 grossing Methods 0.000 claims description 7
- 238000001914 filtration Methods 0.000 claims description 6
- 239000000969 carrier Substances 0.000 claims description 3
- 230000006870 function Effects 0.000 description 5
- 239000000284 extract Substances 0.000 description 4
- 238000004891 communication Methods 0.000 description 3
- 238000013135 deep learning Methods 0.000 description 3
- 238000001514 detection method Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000001228 spectrum Methods 0.000 description 3
- 238000011161 development Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000007476 Maximum Likelihood Methods 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 230000001427 coherent effect Effects 0.000 description 1
- 238000005336 cracking Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012567 pattern recognition method Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 230000003595 spectral effect Effects 0.000 description 1
- 238000010998 test method Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/08—Feature extraction
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Radar Systems Or Details Thereof (AREA)
Abstract
The invention relates to a method and a device for generating an electromagnetic signal classification recognition data set, and an original electromagnetic signal sample is obtained; extracting effective pulses in an original electromagnetic signal sample; quadrature down-conversion of the extracted effective pulse to generate IQ sampling data; and (3) making one-hot codes for the electromagnetic signal data subjected to quadrature down-conversion to form a sample data set. The invention reduces the complexity of the deep network model for electromagnetic signal classification and identification, not only reduces network parameters, but also maintains higher identification accuracy.
Description
Technical Field
The invention relates to a method and a device for generating an electromagnetic signal classification recognition data set, and an electromagnetic signal classification recognition method and a device.
Background
And (3) electromagnetic signal classification and identification, namely extracting characteristics of the electromagnetic signals through a machine learning method and automatically identifying the types of the electromagnetic signals. Electromagnetic signal recognition technology plays an important role in modern wireless communication. In the civil aspect, the method is mainly used in the fields of spectrum detection, dynamic spectrum planning and the like, and aims to ensure the safe work of a communication system and improve the spectrum utilization rate. In the military aspect, the electronic counter-force detector is mainly used for disturbing and cracking enemy signals, so that the purposes of electronic counter-force and counter-force are achieved. However, with the rapid development of communication technology, electromagnetic signals are increasingly diversified, electromagnetic environments are increasingly complex, and difficulty in recognizing the electromagnetic signals is continuously increasing.
Electromagnetic signal classification and identification traditional algorithms are mainly divided into two main categories: a maximum likelihood hypothesis test method based on decision theory and a pattern recognition method based on feature extraction. The former is based on Bayesian theory, applies probability statistics and hypothesis test theory in mathematics, and completes electromagnetic signal classification and identification by combining signals; the latter needs to extract effective characteristic parameters from the signals, and then compares the characteristic parameters with a characteristic library to finish the classification and identification of electromagnetic signals. Both methods have major limitations due to the need for more a priori knowledge and low signal to noise ratio environments. With the development of machine learning theory and method, researchers start to identify models based on data-driven electromagnetic signal classification, the basic idea is to automatically extract signal features from sample data by constructing a deep learning network model, and then train the model under supervision of a loss function to complete electromagnetic signal classification identification.
At present, two processing methods exist for original electromagnetic signal data based on an electromagnetic signal classification and identification method of deep learning: firstly, converting electromagnetic signal data into images, and converting signal recognition problems into target detection problems in the field of image recognition, wherein the method is simple and effective, but greatly improves the complexity and the calculated amount of a network; secondly, the original electromagnetic signals are directly input into a network model, and the identification accuracy is required to be improved although the network parameters are greatly reduced by the method. There is therefore a need for a method of electromagnetic data processing that both reduces network parameters and maintains high recognition accuracy.
Disclosure of Invention
The invention aims to provide a method and a device for generating an electromagnetic signal classification recognition data set, which can reduce network parameters and maintain high recognition accuracy.
Based on the same inventive concept, the invention has four independent technical schemes:
1. a method for generating an electromagnetic signal classification identification dataset comprising the steps of:
step 1: acquiring an original electromagnetic signal sample;
step 2: extracting effective pulses in an original electromagnetic signal sample;
step 3: quadrature down-conversion of the extracted effective pulse to generate IQ sampling data;
step 4: making one-hot codes for electromagnetic signal data after quadrature down-conversion to form a sample data set;
wherein, the step 2 comprises the following steps:
step 2.1: smoothing the background noise of the original electromagnetic signal by using a moving average filter;
step 2.2: the amplitude average value b of the electromagnetic signal after passing through the moving average filter is obtained and used as a threshold value for judging the rising edge and the falling edge of the pulse;
step 2.3: correcting the amplitude average value b according to the burst sample number of each radar signal radiation source until the pulse number is close to the burst sample number of each radar signal radiation source;
step 2.4: and calculating the pulse width average value of all pulse signals, determining a pulse width threshold value according to the pulse width average value, and determining the pulse as a valid pulse if the pulse width of the pulse signal is larger than the pulse width threshold value.
Further, in step 1, the samples are electromagnetic signals generated by a plurality of radar signal radiation sources, each radar signal radiation source has a plurality of burst samples, the carrier frequency is 50MHz, and the sampling rate is 200MSps.
Further, in step 1, the samples are electromagnetic signals generated by 40 radar signal radiation sources, and each radar signal radiation source has 1000 burst samples.
Further, in step 2, the effective pulse in the original electromagnetic signal sample is extracted by the following method,
step 2.1: smoothing the background noise of the original electromagnetic signal by using a moving average filter;
step 2.2: the amplitude average value b of the electromagnetic signal after passing through the moving average filter is obtained and used as a threshold value for judging the rising edge and the falling edge of the pulse;
step 2.3: correcting the amplitude average value b according to the burst sample number of each radar signal radiation source until the pulse number is close to the burst sample number of each radar signal radiation source;
step 2.4: and calculating the pulse width average value of all pulse signals, determining a pulse width threshold value according to the pulse width average value, and determining the pulse as a valid pulse if the pulse width of the pulse signal is larger than the pulse width threshold value.
Further, in step 2.1, smoothing the background noise of the original electromagnetic signal with a moving average filter is achieved by,
a data buffer zone with the size of n is established by a moving average filter, n pieces of sampling data are sequentially stored, the sampling data are the amplitude of electromagnetic signals, the original data are lost every time new data are acquired, and the arithmetic average value of the n pieces of data including the new data is calculated and is recorded as a [ i ].
Further, in step 2.2, if a [ i ]. Ltoreq.b and a [ i+1 ]. Gtoreq.b, i is an index of a rising edge, and if a [ i ]. Gtoreq.b and a [ i+1 ]. Ltoreq.b, i is an index of a falling edge.
Further, in step 2.4, the pulse width threshold C is calculated by the following formula, and the pulse width average value x 0.6=c.
Further, in step 2.1, n takes a value of 40.
Further, in step 3, quadrature down-conversion of the extracted effective pulse is achieved by generating IQ sample data,
and removing intermediate frequency carriers in the effective pulse signals through digital down conversion, extracting an in-phase part and a quadrature part of the signals, and then carrying out low-pass filtering on the in-phase part and the quadrature part of the extracted signals to generate IQ sampling data.
2. An apparatus for generating an electromagnetic signal classification identification dataset for performing the method as described above.
3. The electromagnetic signal classification and identification method utilizes the method for generating the electromagnetic signal classification and identification data set to generate a sample data set, automatically extracts signal characteristics from the sample data set, and trains a model under supervision of a loss function to finish electromagnetic signal classification and identification.
4. An electromagnetic signal classification and identification apparatus, the apparatus being configured to: the method for generating the electromagnetic signal classification recognition data set is utilized to generate a sample data set, signal features are automatically extracted from the sample data set, and then a model is trained under supervision of a loss function, so that electromagnetic signal classification recognition is completed.
The invention has the beneficial effects that:
the invention extracts effective pulse in original electromagnetic signal sample; quadrature down-conversion of the extracted effective pulse to generate IQ sampling data; and (3) making one-hot codes for the electromagnetic signal data subjected to quadrature down-conversion to form a sample data set. The invention reduces the complexity of the deep network model for electromagnetic signal classification and identification, not only reduces network parameters, but also maintains higher identification accuracy. The recognition effect is shown in fig. 1. The average recognition accuracy is 89.2% when the signal-to-noise ratio of the network is greater than 0dB on the down-converted I/Q signal data set, and is 73.5% when the signal-to-noise ratio of the network is greater than 0dB on the original electromagnetic signal data set, so that the improvement effect is 15.7%.
The method comprises the steps of extracting effective pulses from an original electromagnetic signal sample, and smoothing the background noise of the original electromagnetic signal by using a moving average filter; the amplitude average value b of the electromagnetic signal after passing through the moving average filter is obtained and used as a threshold value for judging the rising edge and the falling edge of the pulse; correcting the amplitude average value b according to the burst sample number of each radar signal radiation source until the pulse number is close to the burst sample number of each radar signal radiation source; and calculating the pulse width average value of all pulse signals, determining a pulse width threshold value according to the pulse width average value, and determining the pulse as a valid pulse if the pulse width of the pulse signal is larger than the pulse width threshold value. The invention realizes the smoothing of the background noise of the original electromagnetic signal by utilizing a moving average filter, establishes a data buffer zone with the size of n by utilizing the moving average filter, sequentially stores n pieces of sampling data, wherein the sampling data is the amplitude of the electromagnetic signal, and loses the original data when new data is acquired, and then calculates the arithmetic average value of the n pieces of data including the new data, and is recorded as a [ i ]. If a [ i ] is less than or equal to b and a [ i+1] is more than or equal to b, i is an index of a rising edge, and if a [ i ] is more than or equal to b and a [ i+1] is less than or equal to b, i is an index of a falling edge. The pulse width threshold C is calculated by the following formula, and the pulse width average value x 0.6=c. The invention obtains effective pulse through the method, and further effectively ensures that the invention can reduce network parameters and maintain higher identification accuracy.
The invention realizes quadrature down-conversion of the extracted effective pulse to generate IQ sampling data, removes intermediate frequency carrier wave in the effective pulse signal through digital down-conversion, extracts in-phase part and quadrature part of the signal, and carries out low-pass filtering on the in-phase part and quadrature part of the extracted signal to generate IQ sampling data. The IQ sampling data is generated by the method, so that the network parameters can be reduced, and the high recognition accuracy can be maintained.
Drawings
FIG. 1 is a diagram showing the effect of the present invention;
FIG. 2 is a time domain view of the original electromagnetic signal of the present invention;
FIG. 3 is a graph of effective pulse segmentation of electromagnetic signals according to the present invention;
FIG. 4 is a flow chart of implementing quadrature down-conversion by the electromagnetic signal low-pass filtering method of the present invention;
FIG. 5 is a diagram of the One-hot encoding of the dataset of the present invention.
Detailed Description
The present invention will be described in detail below with reference to the embodiments shown in the drawings, but it should be understood that the embodiments are not limited to the present invention, and functional, method, or structural equivalents and alternatives according to the embodiments are within the scope of protection of the present invention by those skilled in the art.
Example 1
Method for generating electromagnetic signal classification recognition data sets
The method for generating the electromagnetic signal classification recognition data set comprises the following steps:
step 1: a raw electromagnetic signal sample is obtained.
As shown in fig. 2, the samples are electromagnetic signals generated by a plurality of radar signal radiation sources, each radar signal radiation source has a plurality of burst samples, the carrier frequency is 50MHz, and the sampling rate is 200MSps. In this embodiment, the samples are electromagnetic signals generated by 40 radar signal radiation sources, each radar signal radiation source has 1000 burst samples, the carrier frequency is 50MHz, and the sampling rate is 200MSps.
Step 2: valid pulses in the original electromagnetic signal samples are extracted.
Step 2.1: the background noise of the original electromagnetic signal is smoothed using a moving average filter.
A data buffer zone with the size of n is established by a moving average filter, n pieces of sampling data are sequentially stored, the sampling data are the amplitude of electromagnetic signals, the original data are lost every time new data are acquired, and the arithmetic average value of the n pieces of data including the new data is calculated and is recorded as a [ i ]. In this embodiment, n takes a value of 40.
Step 2.2: the average value b of the amplitude of the electromagnetic signal after passing through the moving average filter is obtained and used as a threshold value for judging the rising edge and the falling edge of the pulse.
If a [ i ] is less than or equal to b and a [ i+1] is more than or equal to b, i is an index of a rising edge, and if a [ i ] is more than or equal to b and a [ i+1] is less than or equal to b, i is an index of a falling edge. The indices of rising and falling edges are recorded and the index (i.e., location) in the original electromagnetic signal data.
Step 2.3: and correcting the amplitude average value b according to the burst sample number of each radar signal radiation source until the pulse number is close to the burst sample number of each radar signal radiation source.
In this embodiment, it is known that there are 1000 burst samples per radar signal radiation source, so the number of rising and falling edges should be 1000, and the length of the moving average filter, and the magnitude of the threshold b, is modified according to the number of rising and falling edges obtained in the previous step until the number of rising and falling edges approaches 1000. And then dividing each pulse signal according to the stored rising edge and falling edge index positions.
Step 2.4: and (3) calculating the average pulse width value of all pulse signals, determining a pulse width threshold value according to the average pulse width value, and determining the pulse as a valid pulse if the pulse width of the pulse signal is larger than the pulse width threshold value, as shown in fig. 3.
The pulse width threshold C is calculated by the following formula, and the pulse width average value x 0.6=c.
Step 3: quadrature down-conversion of the extracted effective pulse to generate IQ sampling data;
and removing intermediate frequency carriers in the effective pulse signals through digital down conversion, extracting an in-phase part and a quadrature part of the signals, and then carrying out low-pass filtering on the in-phase part and the quadrature part of the extracted signals to generate IQ sampling data.
As shown in fig. 4, the task of digital down-conversion is to remove the intermediate frequency carrier from the received signal, and to extract the in-phase and quadrature parts of the signal without loss. And the results of the coherent detection of the two branches are subjected to low-pass filtering to obtain the results of the in-phase part and the quadrature part of the signals.
The intermediate frequency signal is first passed through an ADC anti-aliasing filter:
；
the real signal is branched by an in-phase branch multiplier:
；
the real signal is branched by the orthogonal branch multiplier:
；
passing the in-phase output signal through a low pass filter to obtain a signal in-phase part I:
；
passing the quadrature output signal through a low pass filter to obtain a signal quadrature part Q:
；
is a digital real signal generated after the intermediate frequency signal IF passes through an ADC analog-to-digital converter. />Is the amplitude of the signal, +.>For letterCarrier frequency of number->For the phase of the signal, +.>Output signal of in-phase branch,/>Is the output signal of the quadrature branch.
Because IQ sampling requires a lower sampling rate, decimation, which is a process of reducing the sampling rate, is also typically performed to avoid spectral aliasing.
Step 4: and (3) making one-hot codes for the electromagnetic signal data subjected to quadrature down-conversion to form a sample data set, as shown in fig. 5.
The obtained sample data set can be directly input into a constructed deep learning network model for electromagnetic signal classification and identification.
Example two
Device for generating electromagnetic signal classification recognition data set
The apparatus is for performing the method of embodiment one.
Example III
Electromagnetic signal classification and identification method
The method for generating an electromagnetic signal classification recognition dataset of embodiment one is utilized to generate a sample dataset, signal features are automatically extracted from the sample dataset, and then a model is trained under supervision of a loss function to complete electromagnetic signal classification recognition.
Example IV
Electromagnetic signal classifying and identifying device
For performing the following operations: the method for generating an electromagnetic signal classification recognition dataset of embodiment one is utilized to generate a sample dataset, signal features are automatically extracted from the sample dataset, and then a model is trained under supervision of a loss function to complete electromagnetic signal classification recognition.
The above list of detailed descriptions is only specific to practical embodiments of the present invention, and they are not intended to limit the scope of the present invention, and all equivalent embodiments or modifications that do not depart from the spirit of the present invention should be included in the scope of the present invention.
It will be evident to those skilled in the art that the invention is not limited to the details of the foregoing illustrative embodiments, and that the present invention may be embodied in other specific forms without departing from the spirit or essential characteristics thereof. The present embodiments are therefore to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Therefore, the protection scope of the invention is subject to the protection scope of the claims.
Claims (10)
1. A method for generating an electromagnetic signal classification identification dataset, comprising the steps of:
step 1: acquiring an original electromagnetic signal sample;
step 2: extracting effective pulses in an original electromagnetic signal sample;
step 3: quadrature down-conversion of the extracted effective pulse to generate IQ sampling data;
step 4: making one-hot codes for electromagnetic signal data after quadrature down-conversion to form a sample data set;
wherein, the step 2 comprises the following steps:
step 2.1: smoothing the background noise of the original electromagnetic signal by using a moving average filter;
step 2.2: the amplitude average value b of the electromagnetic signal after passing through the moving average filter is obtained and used as a threshold value for judging the rising edge and the falling edge of the pulse;
step 2.3: correcting the amplitude average value b according to the burst sample number of each radar signal radiation source until the pulse number is close to the burst sample number of each radar signal radiation source;
step 2.4: and calculating the pulse width average value of all pulse signals, determining a pulse width threshold value according to the pulse width average value, and determining the pulse as a valid pulse if the pulse width of the pulse signal is larger than the pulse width threshold value.
2. The method for generating an electromagnetic signal classification recognition dataset of claim 1, wherein: in step 1, the samples are electromagnetic signals generated by a plurality of radar signal radiation sources, each radar signal radiation source has a plurality of burst samples, the carrier frequency is 50MHz, and the sampling rate is 200MSps.
3. The method for generating an electromagnetic signal classification recognition dataset of claim 2, wherein: in step 1, the samples are electromagnetic signals generated by 40 radar signal radiation sources, and each radar signal radiation source has 1000 burst samples.
4. The method for generating an electromagnetic signal classification recognition dataset of claim 1, wherein: in step 2.1, smoothing the background noise of the original electromagnetic signal with a moving average filter is achieved by,
a data buffer zone with the size of n is established by a moving average filter, n pieces of sampling data are sequentially stored, the sampling data are the amplitude of electromagnetic signals, the original data are lost every time new data are acquired, and the arithmetic average value of the n pieces of data including the new data is calculated and is recorded as a [ i ].
5. The method for generating an electromagnetic signal classification recognition dataset of claim 4, wherein: in step 2.2, if a [ i ] is not more than b and a [ i+1] is not less than b, i is an index of a rising edge, and if a [ i ] is not less than b and a [ i+1] is not more than b, i is an index of a falling edge.
6. The method for generating an electromagnetic signal classification recognition dataset of claim 1, wherein: in step 2.4, the pulse width threshold C is calculated by the following formula,
pulse width average value x 0.6=c.
7. The method for generating an electromagnetic signal classification recognition dataset of claim 1, wherein: the step 3 comprises the following steps:
and removing intermediate frequency carriers in the effective pulse signals through digital down conversion, extracting an in-phase part and a quadrature part of the signals, and then carrying out low-pass filtering on the in-phase part and the quadrature part of the extracted signals to generate IQ sampling data.
8. An apparatus for generating an electromagnetic signal classification identification dataset, characterized by: for performing the method of any of claims 1 to 7.
9. An electromagnetic signal classification and identification method is characterized in that: a sample dataset is generated using the method of generating an electromagnetic signal classification recognition dataset of any of claims 1 to 7, signal features are automatically extracted from the sample dataset, and then a model is trained under supervision of a loss function to complete electromagnetic signal classification recognition.
10. An electromagnetic signal classification and identification apparatus, characterized in that the apparatus is adapted to perform the following operations: a sample dataset is generated using the method of generating an electromagnetic signal classification recognition dataset of any of claims 1 to 7, signal features are automatically extracted from the sample dataset, and then a model is trained under supervision of a loss function to complete electromagnetic signal classification recognition.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311253730.8A CN116992352B (en) | 2023-09-27 | 2023-09-27 | Method for generating electromagnetic signal classification recognition data set, recognition method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311253730.8A CN116992352B (en) | 2023-09-27 | 2023-09-27 | Method for generating electromagnetic signal classification recognition data set, recognition method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116992352A CN116992352A (en) | 2023-11-03 |
| CN116992352B true CN116992352B (en) | 2023-12-05 |
Family
ID=88532530
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311253730.8A Active CN116992352B (en) | 2023-09-27 | 2023-09-27 | Method for generating electromagnetic signal classification recognition data set, recognition method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116992352B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107192892A (en) * | 2017-06-16 | 2017-09-22 | 国网电力科学研究院武汉南瑞有限责任公司 | The automatic triggering method of thunder alignment system based on lightning electromagnetic signal identification technology |
| CN110868918A (en) * | 2017-07-10 | 2020-03-06 | 健康与环境慕尼黑德国研究中心赫姆霍茨中心(有限公司) | Apparatus and method for photoacoustic sensing |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10830880B2 (en) * | 2018-03-20 | 2020-11-10 | Panosense Inc. | Selecting LIDAR pulse detector depending on pulse type |
| US10830881B2 (en) * | 2018-03-20 | 2020-11-10 | Panosense Inc. | Active signal detection using adaptive identification of a noise floor |
-
2023
- 2023-09-27 CN CN202311253730.8A patent/CN116992352B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107192892A (en) * | 2017-06-16 | 2017-09-22 | 国网电力科学研究院武汉南瑞有限责任公司 | The automatic triggering method of thunder alignment system based on lightning electromagnetic signal identification technology |
| CN110868918A (en) * | 2017-07-10 | 2020-03-06 | 健康与环境慕尼黑德国研究中心赫姆霍茨中心(有限公司) | Apparatus and method for photoacoustic sensing |
Non-Patent Citations (3)
| Title |
|---|
| A machine learning integrated 5.8-GHz continuous-wave radar for honeybee monitoring and behavior classification;Nawaf Aldabashi等;《IEEE Transactions on Microwave RTheory and Techniques》;第71卷(第9期);全文 * |
| 基于EMD-DISPO的Mie散射激光雷达回波信号去噪方法研究;张毅坤;麻晓畅;华灯鑫;陈浩;刘才轩;;光谱学与光谱分析(第11期);全文 * |
| 外场辐射发射测量中瞬变环境信号的抑制;李鹏;周忠元;盛明杰;周祺;胡鹏;;东南大学学报(自然科学版)(第03期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116992352A (en) | 2023-11-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| KR101910540B1 (en) | Apparatus and method for recognizing radar waveform using time-frequency analysis and neural network | |
| CN104052701B (en) | A kind of intrapulse modulation characteristic extract real-time and categorizing system realized based on FPGA | |
| CN107392123B (en) | Radio frequency fingerprint feature extraction and identification method based on coherent accumulation noise elimination | |
| CN111722188A (en) | PRI (pulse repetition index) conversion radar signal sorting method based on STFT (space time Fourier transform) pre-sorting | |
| CN112560803A (en) | Radar signal modulation identification method based on time-frequency analysis and machine learning | |
| CN105429719B (en) | Based on power spectrum and multi-scale wavelet transformation analysis high reject signal detection method | |
| Zhao et al. | 2-D TFPF based on Contourlet transform for seismic random noise attenuation | |
| CN108444954B (en) | Spectral signal peak detection method, device and system | |
| CN105071830B (en) | A kind of detection recognition method of direct sequence signal | |
| CN108647693B (en) | Sea surface infrared target detection method based on binary significance characteristics | |
| CN110532825A (en) | A kind of bar code identifying device and method based on artificial intelligence target detection | |
| CN111680737A (en) | Radar radiation source individual identification method under differential signal-to-noise ratio condition | |
| CN114025379A (en) | Broadband multi-signal detection method, device and equipment | |
| CN111046791A (en) | Current signal filtering and denoising method based on generalized S transform containing variable factors | |
| CN116992352B (en) | Method for generating electromagnetic signal classification recognition data set, recognition method and device | |
| CN114296032A (en) | Real-time complex signal identification and parameter estimation method based on FPGA | |
| CN112751633B (en) | Broadband spectrum detection method based on multi-scale window sliding | |
| CN113920255A (en) | High-efficient mapping system based on point cloud data | |
| CN117111016B (en) | Real-time pulse analysis method and system based on channelization in complex electromagnetic environment | |
| CN113723244A (en) | Radar radiation source signal separation method based on improved variational modal decomposition | |
| CN110991378A (en) | Power amplifier individual identification method and device | |
| CN113848391B (en) | Pulse signal detection and extraction method | |
| CN107341519B (en) | Support vector machine identification optimization method based on multi-resolution analysis | |
| CN115345216A (en) | FMCW radar interference elimination method fusing prior information | |
| CN110532826A (en) | A kind of bar code identifying device and method based on artificial intelligence semantic segmentation |
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 |