CN111988252A - Signal modulation mode identification method based on deep learning - Google Patents
Signal modulation mode identification method based on deep learning Download PDFInfo
- Publication number
- CN111988252A CN111988252A CN202010857048.XA CN202010857048A CN111988252A CN 111988252 A CN111988252 A CN 111988252A CN 202010857048 A CN202010857048 A CN 202010857048A CN 111988252 A CN111988252 A CN 111988252A
- Authority
- CN
- China
- Prior art keywords
- modulation
- signal
- neural network
- convolutional neural
- signals
- 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
- 238000000034 method Methods 0.000 title claims abstract description 21
- 238000013135 deep learning Methods 0.000 title claims abstract description 11
- 238000012549 training Methods 0.000 claims abstract description 27
- 238000013527 convolutional neural network Methods 0.000 claims abstract description 26
- 238000007781 pre-processing Methods 0.000 claims abstract description 5
- 238000010606 normalization Methods 0.000 claims description 4
- 238000004891 communication Methods 0.000 description 6
- 230000005540 biological transmission Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 238000013461 design Methods 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 238000001228 spectrum Methods 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000013136 deep learning model Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L27/00—Modulated-carrier systems
- H04L27/0012—Modulated-carrier systems arrangements for identifying the type of modulation
-
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/40—Software arrangements specially adapted for pattern recognition, e.g. user interfaces or toolboxes therefor
-
- 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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/12—Classification; Matching
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Biophysics (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Molecular Biology (AREA)
- General Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Biomedical Technology (AREA)
- Health & Medical Sciences (AREA)
- Human Computer Interaction (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Digital Transmission Methods That Use Modulated Carrier Waves (AREA)
Abstract
The invention discloses a signal modulation mode identification method based on deep learning, which comprises the following steps: acquiring original IQ data of modulation signals of different modulation modes and carrying out data preprocessing; constructing a convolutional neural network model; constructing a training tool and training the convolutional neural network model to obtain a modulation recognition model; and predicting the input signal, and outputting the modulation mode or modulation protocol of the signal or the type of the self-defined signal. The invention supports effective identification of different modulation modes, different modulation protocols and user-defined signals; the system has the capability of identifying new signals and supports the dynamic expansion of users according to the needs.
Description
Technical Field
The invention relates to the technical field of wireless communication and electromagnetic spectrum supervision, in particular to a signal modulation mode identification technology, and specifically relates to a signal modulation mode identification method based on deep learning.
Background
The demodulation of a communication signal is premised on determining the modulation scheme and parameters of the communication signal, such as signal frequency, signal bandwidth, etc. The modulation mode is one of the main characteristics for distinguishing different modulation signals, after the detection signal and the estimation parameter are carried out, the received modulation signal is correspondingly processed, the judgment of the corresponding modulation mode is completed according to the mode judgment criterion, necessary information is provided for subsequent signal demodulation, implementation of electronic interference, electromagnetic spectrum detection, electronic countermeasure, abnormal signal identification and other non-cooperative communication tasks, and the automatic identification of the modulation mode of the signal is widely applied in the military and civil fields, including wireless communication, navigation, radar and the like.
In order to meet various service requirements, researchers have designed various signal modulation schemes for this purpose. Different channels need to adopt corresponding modulation modes to meet different channel condition requirements. The higher requirements for information transmission speed, transmission bandwidth and transmission quality have pushed the modern communication technology to move forward and the generation of more kinds of modulation schemes, so the demand for signal modulation scheme identification technology is increasing.
In a traditional signal modulation mode identification algorithm based on characteristics, most of the characteristic is realized by manually designing expert characteristics and then extracting and identifying the characteristics, so that a large amount of calculation is needed during signal preprocessing, the robustness is poor, when a new modulation mode is identified, a signal processing expert in the related field is needed to specially design a set of new characteristics for the new characteristic, the whole process is very complex and needs to consume a large amount of manpower and time, after the expert characteristics are successfully extracted, a set of characteristic-based signal modulation identification method needs to be elaborately designed for the new characteristic, and the whole processing link needs an expert in signal processing to carry out precise design and experiment, so that the algorithm is very complicated and complex.
In the prior art, deep learning is also applied to the field of modulation identification, but the modulation type is solidified, a newly added modulation signal cannot be identified, and the identification of a fixed modulation protocol and a user-defined signal type is not supported.
Disclosure of Invention
The invention aims to provide a signal modulation mode identification method based on deep learning, which is used for solving the problems that the modulation signal identification mode in the prior art is complicated, and the modulation type is solidified in the traditional modulation mode identification method based on deep learning, and the modulation protocol and the user-defined signal type cannot be identified.
The invention solves the problems through the following technical scheme:
a signal modulation mode identification method based on deep learning comprises the following steps:
step S1: acquiring enough original IQ data of modulation signals with different modulation modes;
step S2: carrying out data preprocessing on the original IQ data;
step S3: constructing a convolutional neural network model, inputting the preprocessed data into the convolutional neural network model, and after full iterative training, the convolutional neural network model has the capability of identifying signals of different modulation modes;
step S4: the method comprises the following steps of constructing a training tool and training a convolutional neural network model to obtain a modulation recognition model, wherein the constructed training tool is used for assisting to train new signal data by self, and finally has the recognition capability of a new signal, and specifically comprises the following steps:
step S41: building a GUI tool for model training by using a tool kit pyqt, collecting signals of different modulation protocols and/or user-defined types, inputting the signals into a convolutional neural network model, and adjusting the number of nodes of a model output layer by the convolutional neural network model according to the type of the input signals;
step S42: putting the training process of the convolutional neural network model into a GUI tool, and outputting a model file after the training is finished;
step S5: and loading a model file, predicting an input signal, and outputting a modulation mode or a modulation protocol of the signal or a user-defined signal type.
The modulation recognition model generated after training through the training tool can effectively recognize different signal types and different modulation protocols, and has the capability of recognizing new signals, so that a user can dynamically expand according to needs.
The step S2 specifically includes: the raw IQ data is quantized to between-1 and 1 using numerical normalization. The method aims to solve the problem of the numerical scale of data under different modulation signals, avoid the model from being interfered by the numerical scale, and enhance the convergence degree and the modulation mode identification performance of the model.
Compared with the prior art, the invention has the following advantages and beneficial effects:
(1) the invention supports effective identification of different modulation modes, different modulation protocols and user-defined signals; the system has the capability of identifying new signals and supports the dynamic expansion of users according to the needs.
(2) The original IQ data are quantized to be between-1 and 1 by adopting numerical normalization, the problem of numerical scale of the data under different modulation signals is solved, the model is prevented from being interfered by the numerical scale, and the convergence degree of the model and the identification performance of the modulation mode are enhanced.
Drawings
FIG. 1 is a flow chart of the present invention;
FIG. 2 is a schematic diagram of a convolutional neural network model structure in the present invention.
Detailed Description
The present invention will be described in further detail with reference to examples, but the embodiments of the present invention are not limited thereto.
Example (b):
referring to fig. 1, a method for identifying a signal modulation mode based on deep learning includes:
step S1: acquiring enough original IQ data of modulation signals with different modulation modes;
step S2: carrying out data preprocessing on the original IQ data: quantizing the original IQ data to be between-1 and 1 by adopting numerical value normalization;
step S3: constructing a convolutional neural network model (namely a deep learning model) as shown in fig. 2, inputting the preprocessed data into the convolutional neural network model, and performing sufficient iterative training, wherein the convolutional neural network model has the capability of identifying signals of different modulation modes;
step S4: the method comprises the following steps of constructing a training tool and training a convolutional neural network model to obtain a modulation recognition model, wherein the constructed training tool is used for assisting to train new signal data by self, and finally has the recognition capability of a new signal, and specifically comprises the following steps:
step S41: building a GUI tool for model training by using a tool kit pyqt, collecting signals of different modulation protocols and/or user-defined types, inputting the signals into a convolutional neural network model, and adjusting the number of nodes of a model output layer by the convolutional neural network model according to the type of the input signals; the node number of the output layer is changed according to the identification signal type, namely the node number of the output layer is equal to the identification signal type;
step S42: putting the training process of the convolutional neural network model into a GUI tool, and outputting a model file after the training is finished;
step S5: and loading a model file, predicting an input signal, and outputting a modulation mode or a modulation protocol of the signal or a user-defined signal type, namely an identification result.
The convolutional neural network is introduced to classify different signal modulation modes, the high abstraction and the representation learning capacity of the convolutional neural network are fully utilized, the local information and the detail information of the modulation signals are extracted, the high-frequency characteristic and the low-frequency characteristic of the signals are obtained through layer-by-layer analysis of the multi-convolutional layer, and finally the high-level characteristic which can be used for classification is obtained through abstraction.
The method comprises the steps of collecting original IQ data of different modulation modes, inputting the data into a convolutional neural network, and identifying and classifying modulation protocols of signals after full iterative training, wherein the data has high precision;
the method comprises the steps of collecting original IQ data and new signal data of signals with different modulation protocols, inputting a model obtained by training a convolutional neural network through a training tool, and after full iterative training, not only identifying the modulation protocols, but also adapting to a modulation mode of the new signals, and having a high identification rate, thereby solving the problem that the traditional modulation identification method is difficult to expand the new signals.
Although the present invention has been described herein with reference to the illustrated embodiments thereof, which are intended to be preferred embodiments of the present invention, it is to be understood that the invention is not limited thereto, and that numerous other modifications and embodiments can be devised by those skilled in the art that will fall within the spirit and scope of the principles of this disclosure.
Claims (2)
1. A signal modulation mode identification method based on deep learning is characterized by comprising the following steps:
step S1: acquiring original IQ data of modulation signals of different modulation modes;
step S2: carrying out data preprocessing on the original IQ data;
step S3: constructing a convolutional neural network model;
step S4: constructing a training tool and training a convolutional neural network model to obtain a modulation recognition model, which specifically comprises the following steps:
step S41: building a GUI tool for model training by using a tool kit pyqt, collecting signals of different modulation protocols and/or user-defined types, inputting the signals into a convolutional neural network model, and adjusting the number of nodes of a model output layer by the convolutional neural network model according to the type of the input signals;
step S42: putting the training process of the convolutional neural network model into a GUI tool, and outputting a model file after the training is finished;
step S5: and loading a model file, predicting an input signal, and outputting a modulation mode or a modulation protocol of the signal or a user-defined signal type.
2. The method for identifying a signal modulation scheme based on deep learning according to claim 1, wherein the step S2 specifically comprises: the raw IQ data is quantized to between-1 and 1 using numerical normalization.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010857048.XA CN111988252A (en) | 2020-08-24 | 2020-08-24 | Signal modulation mode identification method based on deep learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010857048.XA CN111988252A (en) | 2020-08-24 | 2020-08-24 | Signal modulation mode identification method based on deep learning |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111988252A true CN111988252A (en) | 2020-11-24 |
Family
ID=73443808
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010857048.XA Pending CN111988252A (en) | 2020-08-24 | 2020-08-24 | Signal modulation mode identification method based on deep learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111988252A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115277324A (en) * | 2022-07-25 | 2022-11-01 | 电信科学技术第五研究所有限公司 | FSK signal identification method based on convolutional neural network |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108616470A (en) * | 2018-03-26 | 2018-10-02 | 天津大学 | Modulation Signals Recognition method based on convolutional neural networks |
| CN109787927A (en) * | 2019-01-03 | 2019-05-21 | 荆门博谦信息科技有限公司 | Modulation Identification method and apparatus based on deep learning |
| CN109922019A (en) * | 2019-02-26 | 2019-06-21 | 天津大学 | Intelligent communication method based on deep learning |
| CN110728230A (en) * | 2019-10-10 | 2020-01-24 | 江南大学 | Signal modulation mode identification method based on convolution limited Boltzmann machine |
| CN110855591A (en) * | 2019-12-09 | 2020-02-28 | 山东大学 | QAM and PSK signal intra-class modulation classification method based on convolutional neural network structure |
-
2020
- 2020-08-24 CN CN202010857048.XA patent/CN111988252A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108616470A (en) * | 2018-03-26 | 2018-10-02 | 天津大学 | Modulation Signals Recognition method based on convolutional neural networks |
| CN109787927A (en) * | 2019-01-03 | 2019-05-21 | 荆门博谦信息科技有限公司 | Modulation Identification method and apparatus based on deep learning |
| CN109922019A (en) * | 2019-02-26 | 2019-06-21 | 天津大学 | Intelligent communication method based on deep learning |
| CN110728230A (en) * | 2019-10-10 | 2020-01-24 | 江南大学 | Signal modulation mode identification method based on convolution limited Boltzmann machine |
| CN110855591A (en) * | 2019-12-09 | 2020-02-28 | 山东大学 | QAM and PSK signal intra-class modulation classification method based on convolutional neural network structure |
Non-Patent Citations (1)
| Title |
|---|
| 白晶等: ""基于深度学习的调制识别研究"", 《中国无线电》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115277324A (en) * | 2022-07-25 | 2022-11-01 | 电信科学技术第五研究所有限公司 | FSK signal identification method based on convolutional neural network |
| CN115277324B (en) * | 2022-07-25 | 2023-11-10 | 电信科学技术第五研究所有限公司 | FSK signal identification method based on convolutional neural network |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107124381B (en) | Automatic identification method for digital communication signal modulation mode | |
| CN109450834B (en) | Communication signal classification and identification method based on multi-feature association and Bayesian network | |
| CN114239749B (en) | Modulation identification method based on residual shrinkage and two-way long-short-term memory network | |
| CN111585922A (en) | Modulation mode identification method based on convolutional neural network | |
| CN110659684A (en) | Convolutional neural network-based STBC signal identification method | |
| CN111988252A (en) | Signal modulation mode identification method based on deep learning | |
| CN111817803A (en) | Frequency spectrum sensing method and system based on correlation coefficient and K-means clustering algorithm and computer readable storage medium | |
| CN110826425A (en) | VHF/UHF frequency band radio signal modulation mode identification method based on deep neural network | |
| CN113095162B (en) | Spectrum sensing method based on semi-supervised deep learning | |
| CN111046697A (en) | Adaptive modulation signal identification method based on fuzzy logic system | |
| CN114024808A (en) | Modulation signal identification method and system based on deep learning | |
| CN113343868A (en) | Radiation source individual identification method and device, terminal and storage medium | |
| CN106656201B (en) | Compression method based on amplitude-frequency characteristics of sampled data | |
| CN115866615B (en) | Wireless network communication relation discovery method based on electromagnetic spectrum characteristics | |
| CN109995690B (en) | Neural network self-optimization method for MFSK digital signal subclass modulation recognition | |
| CN106603976B (en) | Intelligent microwave frequency band radio monitoring control system | |
| CN110880020A (en) | Self-adaptive trans-regional base station energy consumption model migration and compensation method | |
| CN113688953B (en) | Industrial control signal classification method, device and medium based on multilayer GAN network | |
| CN116016071A (en) | Modulation signal identification method based on double-flow fusion CNN-BiLSTM network | |
| CN116017257A (en) | Intelligent production method and system for loudspeaker | |
| CN110784887B (en) | Method for detecting number of abnormal signal sources in gridding radio signal monitoring system | |
| CN111461007A (en) | Automatic modulation signal identification method and device based on fuzzy logic | |
| CN113726686A (en) | Flow identification method and device, electronic equipment and storage medium | |
| CN117611957B (en) | Unsupervised visual representation learning method and system based on unified positive and negative pseudo labels | |
| CN116017604B (en) | Network integrated communication method and system applied to intelligent production equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20201124 |