CN112580598A - Radio signal classification method based on multi-channel Diffpool - Google Patents
Radio signal classification method based on multi-channel Diffpool Download PDFInfo
- Publication number
- CN112580598A CN112580598A CN202011605253.3A CN202011605253A CN112580598A CN 112580598 A CN112580598 A CN 112580598A CN 202011605253 A CN202011605253 A CN 202011605253A CN 112580598 A CN112580598 A CN 112580598A
- Authority
- CN
- China
- Prior art keywords
- channel
- diffpool
- data
- signal
- network
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 53
- 238000012545 processing Methods 0.000 claims abstract description 26
- 239000013598 vector Substances 0.000 claims abstract description 25
- 238000013145 classification model Methods 0.000 claims abstract description 15
- 230000004913 activation Effects 0.000 claims abstract description 10
- 238000007781 pre-processing Methods 0.000 claims abstract description 7
- 230000008569 process Effects 0.000 claims description 20
- 238000010586 diagram Methods 0.000 claims description 16
- 239000011159 matrix material Substances 0.000 claims description 10
- 238000011176 pooling Methods 0.000 claims description 3
- 238000010276 construction Methods 0.000 claims 1
- 238000013507 mapping Methods 0.000 description 9
- 230000006855 networking Effects 0.000 description 7
- 230000000007 visual effect Effects 0.000 description 7
- 230000006870 function Effects 0.000 description 6
- 238000004891 communication Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 3
- 238000000605 extraction Methods 0.000 description 3
- 238000007476 Maximum Likelihood Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000001228 spectrum Methods 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000010801 machine learning 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
- 238000012706 support-vector machine Methods 0.000 description 1
- 238000010998 test method Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
Images
Classifications
-
- 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
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/12—Classification; Matching
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- Software Systems (AREA)
- Molecular Biology (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Mobile Radio Communication Systems (AREA)
- Digital Transmission Methods That Use Modulated Carrier Waves (AREA)
Abstract
A method for classifying radio signals based on multi-channel Diffpool, comprising: acquiring a radio modulation signal data set, preprocessing the radio modulation signal data set, and splitting each signal sample into data of two channels with equal length; constructing two one-dimensional convolutional layers with the same parameter setting but different weights, respectively processing the data of the two channels obtained after preprocessing to obtain the convolved channel data of the two channels, processing the convolved channel data by using a nonlinear activation function to obtain the activated channel data of the two channels, and obtaining two network graphs according to the two activated channel data; modifying a graph classification model Diffpool in the field of network graphs, and respectively processing the two obtained network graphs in two defined channels by using modified models with different initialization weights to obtain two corresponding one-dimensional feature vectors; and splicing the two one-dimensional feature vectors into one-dimensional feature vector, and processing the obtained feature vector by using two full-connection layers to obtain a final classification result.
Description
Technical Field
The invention relates to data mining, network science and data analysis technologies, in particular to a radio signal classification method based on multi-channel Diffpool.
Background
The essence of radio signal classification is to distinguish the modulation patterns to which different signals belong, so that the radio modulated signal is demodulated to obtain the original signal with valid information. Modulation identification of radio signals is an important component of the communication field, and in order to realize rapid and safe transmission of messages by means of a channel under the condition of fully utilizing channel capacity, baseband signals are generally not directly transmitted, but need to be modulated to obtain modulation signals, so that the modulation signals have more effective spectrum utilization rate and higher communication rate. With the rapid development of the current radio communication technology, the electromagnetic environment becomes more complex, and it is increasingly important to correctly and efficiently identify the modulation type to which the modulation signal belongs, which is critical in both civil and military aspects, such as spectrum management, electronic countermeasure, and communication reconnaissance.
In the radio signal classification problem, the conventional classification method includes a modulation recognition method based on feature extraction and a maximum likelihood hypothesis test method. The modulation identification method based on feature extraction needs manual calculation and extraction of a large number of features such as histogram features, statistical moment features, transform domain features and the like, and then classification is carried out by using a classifier in the field of machine learning. In addition, the maximum likelihood hypothesis test rule is to process the likelihood function of the radio signal to obtain statistic for comparison with a threshold, and finally realize modulation signal classification.
With the development of deep learning, people use the convolutional neural network for classification of modulation signals, and a good effect is achieved. In addition, some methods for building a visible graph are proposed to map signal data into a network graph, so as to realize classification and identification of radio signals by a graph classification technology in the field of network graphs.
At present, a method for mapping signals into network maps and then classifying the network maps by a visual map networking method exists, but most methods are complex in process and rigid in mapping process, and the network maps obtained by mapping cannot be guaranteed to better represent original radio signals. For example, patent application No. CN202010566006.0 discloses a radar antenna scanning pattern recognition method based on a limited-penetration visual map, which converts signals into a network map by a visual map networking method LPVG, extracts network characteristic parameters, and classifies the network characteristic parameters by a support vector machine SVM. Compared with the prior art, the method provided by the invention does not need to artificially convert signals and artificially calculate some characteristic parameters of the network diagram, can realize an end-to-end classification framework as long as the original radio signals are split into data of two channels, and can simultaneously and autonomously train the processes of mapping to obtain the network diagram, extracting the characteristic vectors of the network diagram and classifying the characteristic vectors, thereby simplifying the classification process, not requiring the support of corresponding professional knowledge and improving the classification precision.
Disclosure of Invention
The existing visual networking method generally maps signals into a network graph according to a fixed rule, and then classifies the network graph to realize signal classification. The invention aims to overcome the problems of low mapping efficiency, rigid mapping method, low classification precision and the like of the classification method, provides an end-to-end multi-channel Diffpool-based radio signal classification method, divides original radio signal data into data of I and Q channels, and respectively processes the I channel and the Q channel by using two different one-dimensional convolutional layers and a ReLU activation functionChannel data to obtain two corresponding feature matrixes, thereby obtaining two network graphs GIAnd GQThen, two network graph classification models Diffpool with the last layer of full connection layer removed are used to process the network graph G respectivelyIAnd GQAnd corresponding one-dimensional feature vectors are obtained and spliced into one-dimensional feature vector, and then the one-dimensional feature vector is classified by using two full connection layers, so that the classification of the radio signal data is completed, the signal can be flexibly and efficiently mapped into a network diagram through autonomous training, the classification process is simplified, and the classification precision can be improved.
In order to achieve the purpose, the invention provides the following scheme: the invention provides a radio signal classification method based on multi-channel Diffpool, which comprises the following steps:
s1, acquiring a radio modulation signal data set and preprocessing the radio modulation signal data set; the pretreatment process comprises the following steps: each signal sample is split into two channels of data of equal length.
S2, constructing two one-dimensional convolutional layers with the same parameter setting but different weights, respectively processing the data of the two channels obtained after preprocessing to obtain the convolved channel data of the two channels, processing the convolved channel data by using a nonlinear activation function to obtain the activated channel data of the two channels, and obtaining two network graphs according to the two activated channel data;
the number of input channels of the two one-dimensional convolution layers in the step S2 is 1, the number of output channels is 128, the size of a convolution kernel is 3, the step size is 1, and the padding value is 1;
the nonlinear activation function used is ReLU.
S3, modifying a graph classification model Diffpool in the field of network graphs, and respectively processing the two network graphs obtained in the step S2 by using modified models with different initialization weights in the two constructed channels to obtain two corresponding one-dimensional feature vectors;
the specific process of modifying the network graph classification model Diffpool in step S3 is as follows: and removing the last full connection layer for classification in the network map classification model Diffpool, and reserving other network layers such as a map pooling layer to obtain a modified model.
And S4, splicing the two one-dimensional characteristic vectors into one-dimensional characteristic vector, and processing the obtained characteristic vector by using two full-connection layers to obtain a final classification result.
Preferably, the specific way of selecting and preprocessing the radio modulation signal data set in step S1 is:
using a public radio modulation signal data set rml2016.10a, which has 11 modulation types in total, namely CPFSK, PAW4, 8PSK, AM-DSB, AM-SSB, BPSK, GFSK, QAM16, QAM64, QPSK and WBFM, each modulation type has 20 signal-to-noise ratios, each signal-to-noise ratio has 1000 samples, wherein each signal sample has a shape of 2x128, namely L is 128, and each signal sample consists of I-channel data and Q-channel data, and data of an I channel and a Q channel, which are respectively represented as I ═ { I ═ Q channel data, are extracted from the data1，I2，…，In，…，I128Q ═ Q1，Q2，…，Qn，…，Q128}。
Preferably, the one-dimensional convolutional layer Conv1D of step S2IAnd Conv1DQHas an input channel number of 1, an output channel number of 128, a convolution kernel size of 3, a step size of 1, a padding value of 1, and Conv1DIAnd Conv1DQThe initialization weights of the data processing system are different, and the selected nonlinear activation function is ReLU.
Preferably, the specific process of modifying the network graph classification model Diffpool in step S3 is as follows: and removing the last full connection layer for classification in the network map classification model Diffpool, and reserving other network layers such as a map pooling layer and the like.
Preferably, the entire radio classification framework is end-to-end, and the convolutional layers used in the process, the modified Diffpool, and the fully-connected layers can be trained together.
The invention has the following advantages:
the classification method of the invention fully utilizes the self-learning ability of the neural network, leads the signal to be mapped into a suitable network diagram in a learning way through the one-dimensional convolution layer, fully retains the characteristic information implied in the original radio signal in the mapping process, and then modifies the classical model Diffpool used for classifying the network diagram in the field of the network diagram into a model capable of classifying the multi-channel network diagram, finally forms an end-to-end classification frame, avoids the complex operation of manually extracting the characteristics, compared with the visible diagram networking method of converting the signal into the network diagram according to the fixed rule, the method has high efficiency, flexible mapping method, reduced operation complexity and improved classification precision.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings needed to be used in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings without inventive exercise.
FIG. 1 is a schematic flow diagram of the present invention;
fig. 2 is a schematic diagram of the process of radio signal classification by using the one-dimensional convolutional layer and the modified multi-channel Diffpool model of the method of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in further detail below.
As shown in fig. 1, the present invention provides a method for classifying radio signals based on multi-channel Diffpool, comprising the following steps:
and S1, preprocessing each radio modulation signal with the shape of 2xL, wherein each signal sample comprises data of two channels of I and Q, and splitting the signal sample into data of the I channel and the Q channel with the shape of 1 xL.
The present embodiment uses the disclosed radio modulation signal data set rml2016.10a, which collectively has 11 modulation types of CPFSK, PAW4, 8PSK, AM-DSB, AM-SSB, BPSK, GFSK, QAM16, QAM64, QPSK, and WBFM, each modulation type having 20 signal-to-noise ratios, each signal-to-noise ratio having 1000 samples, where each signal sample has a shape of 2 × 128, i.e., L is 128.
Let each signal sample be expressed as:
S＝{(I1，Q1)，(I2，Q2)，…，(In，Qn)，…，(I128，Q128)} (1)
where 128 denotes the length of the signal, i.e. the number of time points, InAnd QnRespectively representing signal values of the I channel and the Q channel corresponding to the nth time point, extracting data of the I channel and the Q channel from the signal values, and respectively representing that I is equal to { I ═ I1，I2，…，In，…，I128Q ═ Q1，Q2，…，Qn，…，Q128}。
S2, constructing two different one-dimensional convolutional layers Conv1DIAnd Conv1DQRespectively processing the I channel data and the Q channel data to obtain corresponding convolution-processed channel data I 'and Q' with the shapes of LxL, processing the I 'channel data and the Q' channel data by using a nonlinear activation function to obtain corresponding activated channel data I 'and Q' with the shapes of LxL, regarding the activated channel data I 'and Q' as characteristic matrixes capable of representing network diagrams, and obtaining corresponding network diagrams GIAnd GQ。
As shown in FIG. 2, this example constructed a one-dimensional convolutional layer Conv1DIWhen processing I channel data with the shape of 1x128, the number of input channels of the one-dimensional convolutional layer is 1, the number of output channels is 128, the size of the convolutional kernel is 3, the step size is 1, the padding value is 1, a matrix I' with the shape of 128x128 can be obtained, and when processing Q channel data, a one-dimensional convolutional layer Conv1D with the same parameter setting but different weight values is also constructedQResulting in a matrix Q' of 128x128 in shape.
Processing the matrix I 'and the matrix Q' by using the nonlinear activation function ReLU to obtain the characteristic matrix I 'and Q' with the shapes of 128x128, and obtaining the corresponding network graph G according to the characteristic matrix I 'and Q'IAnd GQ。
S3, modifying a graph classification model Diffpool in the field of network graphs, constructing two channels, and processing a network graph G in the first channel by using the graph classification model Diffpool in the field of network graphsIDiffpool processing network graph G using different parameters in the second channelQRespectively extracting to obtain a network graph GIAnd GQFeature vector F ofIAnd FQThe shapes are all 1 xK.
When modifying the network map classification model Diffpool, the last full connection layer for classification in the network map classification model Diffpool is removed, and then the network map G is processed by the rest partIObtaining a one-dimensional feature vector F with the length of KIThe network graph G is processed in the same way with Diffpool with different parametersQObtaining a one-dimensional feature vector F with the length of KQ。
S4, as shown in FIG. 2, splicing the feature vectors FIAnd FQAnd obtaining a feature vector F with the shape of 1x2K, processing the feature vector F by using two full-connection layers for classification, wherein the whole radio classification framework is end-to-end, and a convolutional layer, a Diffpool and the full-connection layers used in the process can be trained together.
Compared with the limited-crossing visual graph networking method proposed in the thesis of limited-crossing visual graph networking model, which processes the same data set, the method of the invention can achieve the classification accuracy of 56.57% in the disclosed radio signal data set RML2016.10a, and is superior to the classification accuracy obtained by the visual graph networking method according to the fixed conversion rule. The invention has high efficiency and flexible mapping method, reduces the operation complexity and improves the classification precision of the radio signal data.
The above-described embodiments are merely illustrative of the preferred embodiments of the present invention, and do not limit the scope of the present invention, and various modifications and improvements of the technical solutions of the present invention can be made by those skilled in the art without departing from the spirit of the present invention, and the technical solutions of the present invention are within the scope of the present invention defined by the claims.
Claims (5)
1. A radio signal classification method based on multi-channel Diffpool comprises the following steps:
s1, selecting a radio modulation signal data set as a sample, preprocessing a radio modulation signal, wherein each signal sample is 2xL in shape and comprises data of two channels I and Q, and splitting the signal sample into data of a channel I and a channel Q which are 1xL in shape;
let each signal sample be expressed as:
S＝{(I1,Q1),(I2,Q2),…,(In,Qn),…,(IL,QL)} (1)
wherein L represents the length of the signal, i.e. the number of time points, InAnd QnRespectively representing signal values of the I channel and the Q channel corresponding to the nth time point, extracting data of the I channel and the Q channel from the signal values, and respectively representing that I is equal to { I ═ I1,I2,…,In,…,ILQ ═ Q1,Q2,…,Qn,…,QL}；
Step S2. Using two different one-dimensional convolution layers Conv1DIAnd Conv1DQRespectively processing the I channel data and the Q channel data to obtain corresponding convolution-processed channel data I 'and Q' with the shapes of LxL, processing the I 'channel data and the Q' channel data by using a nonlinear activation function to obtain corresponding activated channel data I 'and Q' with the shapes of LxL, regarding the activated channel data I 'and Q' as feature matrixes capable of representing network diagrams, and obtaining corresponding network diagrams GIAnd GQ；
Construction of a one-dimensional convolutional layer Conv1D for processing I-channel dataIThe number of input channels of the one-dimensional convolutional layer is 1, the number of output channels is L, the size of the convolutional kernel is 3, the step length is 1, and the padding value is 1, which can beTo obtain a matrix I' with the shape of LxL, when processing Q channel data, a one-dimensional convolutional layer Conv1D with the same parameter setting but different weights is constructedQObtaining a matrix Q' with the shape of LxL;
processing the matrix I 'and the matrix Q' by using a nonlinear activation function to obtain characteristic matrices I 'and Q' with shapes of LxL, and obtaining a corresponding network graph G according to the characteristic matrices I 'and Q'IAnd GQ；
S3, modifying a graph classification model Diffpool in the field of network graphs, constructing two channels, and processing a network graph G in the first channel by using the modified DiffpoolIModified Diffpool processing network graph G using different parameters in the second channelQRespectively extracting a network graph GIAnd GQFeature vector F ofIAnd FQThe shapes are all 1 xK;
when modifying the network graph classification model Diffpool, the last full connection layer for classification in Diffpool is removed, and then the network graph G is processed by the rest partIObtaining a one-dimensional feature vector F with the length of KIThe network graph G is processed in the same way with a modified Diffpool of different parametersQObtaining a one-dimensional feature vector F with the length of KQ；
S4, splicing the feature vectors FIAnd FQAnd obtaining a feature vector F with the shape of 1x2K, processing the feature vector F by using two full-connection layers for classification, wherein the convolutional layer, the Diffpool and the full-connection layers used in the process can be trained together.
2. A multi-channel Diffpool-based radio signal classification method as claimed in claim 1, characterized in that: the specific way in which the step S1 selects and pre-processes the radio modulation signal data set is:
using the disclosed radio modulated Signal data set RML2016.10a, the data set has 11 modulation types including CPFSK, PAW4, 8PSK, AM-DSB, AM-SSB, BPSK, GFSK, QAM16, QAM64, QPSK and WBFM, each modulation type has 20 signal-to-noise ratios, each signal-to-noise ratio has 1000 samples, wherein each signal sample has a shape of 2x128, namely LFor 128, each signal sample is composed of I channel data and Q channel data, from which data for I channel and Q channel, respectively denoted I ═ Q, are extracted1,I2,…,In,…,I128Q ═ Q1,Q2,…,Qn,…,Q128}。
3. A multi-channel Diffpool-based radio signal classification method according to claim 1, characterized in that:
the one-dimensional convolutional layer Conv1D of step S2IAnd Conv1DQHas an input channel number of 1, an output channel number of 128, a convolution kernel size of 3, a step size of 1, a padding value of 1, and Conv1DIAnd Conv1DQThe initialization weights of the data processing system are different, and the selected nonlinear activation function is ReLU.
4. A multi-channel Diffpool-based radio signal classification method according to claim 1, characterized in that:
the specific process of modifying the network graph classification model Diffpool in step S3 is as follows: and removing the last full connection layer for classification in the network map classification model Diffpool, and reserving other network layers such as a map pooling layer and the like.
5. A multi-channel Diffpool-based radio signal classification method according to claim 1, characterized in that:
the entire radio classification framework is end-to-end, and the convolutional layers used in the process, the modified Diffpool, and the full connectivity layer can be trained together.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011605253.3A CN112580598B (en) | 2020-12-30 | 2020-12-30 | Radio signal classification method based on multichannel Diffpool |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011605253.3A CN112580598B (en) | 2020-12-30 | 2020-12-30 | Radio signal classification method based on multichannel Diffpool |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112580598A true CN112580598A (en) | 2021-03-30 |
| CN112580598B CN112580598B (en) | 2024-02-13 |
Family
ID=75144227
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011605253.3A Active CN112580598B (en) | 2020-12-30 | 2020-12-30 | Radio signal classification method based on multichannel Diffpool |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112580598B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113392731A (en) * | 2021-05-31 | 2021-09-14 | 浙江工业大学 | Modulated signal classification method and system based on graph neural network |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107979554A (en) * | 2017-11-17 | 2018-05-01 | 西安电子科技大学 | Radio signal Modulation Identification method based on multiple dimensioned convolutional neural networks |
| CN112069883A (en) * | 2020-07-28 | 2020-12-11 | 浙江工业大学 | Deep learning signal classification method fusing one-dimensional and two-dimensional convolutional neural network |
-
2020
- 2020-12-30 CN CN202011605253.3A patent/CN112580598B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107979554A (en) * | 2017-11-17 | 2018-05-01 | 西安电子科技大学 | Radio signal Modulation Identification method based on multiple dimensioned convolutional neural networks |
| CN112069883A (en) * | 2020-07-28 | 2020-12-11 | 浙江工业大学 | Deep learning signal classification method fusing one-dimensional and two-dimensional convolutional neural network |
Non-Patent Citations (1)
| Title |
|---|
| 杨洁;夏卉;: "基于卷积神经网络的通信信号调制识别研究", 计算机测量与控制, no. 07 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113392731A (en) * | 2021-05-31 | 2021-09-14 | 浙江工业大学 | Modulated signal classification method and system based on graph neural network |
| CN113392731B (en) * | 2021-05-31 | 2023-06-23 | 浙江工业大学 | Modulation signal classification method and system based on graph neural network |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112580598B (en) | 2024-02-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107979554B (en) | Radio signal Modulation Identification method based on multiple dimensioned convolutional neural networks | |
| CN108234370B (en) | Communication signal modulation mode identification method based on convolutional neural network | |
| CN107124381B (en) | Automatic identification method for digital communication signal modulation mode | |
| CN110163282B (en) | Modulation mode identification method based on deep learning | |
| CN112702294B (en) | Modulation recognition method for multi-level feature extraction based on deep learning | |
| CN110266620A (en) | 3D MIMO-OFDM system channel estimation method based on convolutional neural networks | |
| CN113014524B (en) | Digital signal modulation identification method based on deep learning | |
| CN112069883B (en) | Deep learning signal classification method integrating one-dimensional two-dimensional convolutional neural network | |
| CN114157539B (en) | Data-aware dual-drive modulation intelligent identification method | |
| CN111461025A (en) | Signal identification method for self-evolving zero-sample learning | |
| CN114663685B (en) | Pedestrian re-recognition model training method, device and equipment | |
| CN110659684A (en) | Convolutional neural network-based STBC signal identification method | |
| Lin et al. | A real-time modulation recognition system based on software-defined radio and multi-skip residual neural network | |
| CN115186712A (en) | Modulated signal identification method and system | |
| CN114398931A (en) | Modulation recognition method and system based on numerical characteristic and image characteristic fusion | |
| CN112580598B (en) | Radio signal classification method based on multichannel Diffpool | |
| CN114595729A (en) | Communication signal modulation identification method based on residual error neural network and meta-learning fusion | |
| CN109274626B (en) | Modulation identification method based on constellation diagram orthogonal scanning characteristics | |
| CN113076925B (en) | M-QAM signal modulation mode identification method based on CNN and ELM | |
| CN114615118A (en) | Modulation identification method based on multi-terminal convolution neural network | |
| CN113902095A (en) | Automatic modulation identification method, device and system for wireless communication | |
| CN113486724A (en) | Modulation identification model based on CNN-LSTM multi-tributary structure and multiple signal representations | |
| CN112565128A (en) | Radio signal modulation recognition network based on hybrid neural network and implementation method | |
| CN115935252A (en) | Communication signal modulation pattern recognition method and device based on self-supervision contrast learning | |
| CN115086123A (en) | Modulation identification method and system based on fusion of time-frequency graph and constellation diagram |
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 |