CN110321953A - Deep learning intelligent modulation recognition methods based on circulation Power estimation - Google Patents
Deep learning intelligent modulation recognition methods based on circulation Power estimation Download PDFInfo
- Publication number
- CN110321953A CN110321953A CN201910594826.8A CN201910594826A CN110321953A CN 110321953 A CN110321953 A CN 110321953A CN 201910594826 A CN201910594826 A CN 201910594826A CN 110321953 A CN110321953 A CN 110321953A
- Authority
- CN
- China
- Prior art keywords
- modulation
- neural network
- deep learning
- cyclic
- spectrum estimation
- 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 39
- 238000013135 deep learning Methods 0.000 title claims abstract description 16
- 125000004122 cyclic group Chemical group 0.000 claims abstract description 54
- 238000001228 spectrum Methods 0.000 claims abstract description 53
- 238000013528 artificial neural network Methods 0.000 claims abstract description 23
- 230000006870 function Effects 0.000 claims abstract description 15
- 230000003595 spectral effect Effects 0.000 claims abstract description 13
- 238000012549 training Methods 0.000 claims abstract description 5
- 238000010586 diagram Methods 0.000 claims description 14
- 238000004364 calculation method Methods 0.000 claims description 9
- 238000009499 grossing Methods 0.000 claims description 4
- 230000000644 propagated effect Effects 0.000 claims description 4
- 230000000087 stabilizing effect Effects 0.000 claims description 3
- 238000000605 extraction Methods 0.000 abstract description 6
- 238000012545 processing Methods 0.000 abstract description 3
- 239000000284 extract Substances 0.000 abstract description 2
- 230000011664 signaling Effects 0.000 abstract 1
- 238000005516 engineering process Methods 0.000 description 5
- 238000004891 communication Methods 0.000 description 4
- 102100024170 Cyclin-C Human genes 0.000 description 3
- 101000980770 Homo sapiens Cyclin-C Proteins 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 3
- 238000012937 correction Methods 0.000 description 3
- 238000007476 Maximum Likelihood Methods 0.000 description 2
- 230000001149 cognitive effect Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000009897 systematic effect Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000003930 cognitive ability Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000007639 printing 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
-
- 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
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)
- General Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
Abstract
Deep learning intelligent modulation recognition methods provided by the invention based on circulation Power estimation, comprising the following steps: modulated signal is generated according to carrier frequency and chip rate；Circulation Power estimation is carried out to modulated signal, extracts the sectional view of circulation spectral function；Using sectional view as feature, training deep neural network；It is identified using modulation system of the deep neural network to unknown signaling.Deep learning intelligent modulation recognition methods provided by the invention based on circulation Power estimation, circulation Power estimation is combined with deep neural network, using the preferable Classification and Identification ability of the intelligent processing capacity and Cyclic Spectrum of neural network, the performance of entire signal modulate system is improved；Merely with the sectional view of circulation spectral function, the step of extraction recycles spectrum signature is eliminated, Algorithms T-cbmplexity is reduced.
Description
Technical Field
The invention relates to the field of mobile communication, in particular to a deep learning intelligent modulation identification method based on cyclic spectrum estimation.
Background
With the development of communication technology, the complexity and unpredictability of wireless communication environment are greatly increased, and cognitive radio is widely applied due to its cognitive ability, ability to recognize channel characteristics, and dynamic adjustment of allocation of system resources. The automatic identification technology of the signal modulation mode is a key technology in the cognitive radio communication system.
The existing communication signal modulation mode identification methods are mainly divided into two categories, namely a method based on maximum likelihood theory and a pattern identification method based on characteristic parameter extraction, and because the method based on maximum likelihood theory has overlarge calculation complexity and has an unsatisfactory identification effect under the condition of low signal-to-noise ratio, the pattern identification method based on characteristic parameter extraction is usually adopted.
In 1982 to 1989, Gardner w.a.w. published several works, giving a relevant definition of cyclostationary properties, and giving algorithms for identifying signals using different classifications of cyclic spectral properties of different modulated signals.
The documents "Han dynasty, Chua and Wujiang xing" modulation analysis and identification spectrum correlation method [ J ] systematic engineering and electronic technology, 2001(03):34-36+46 "propose 6 characteristic functions based on cyclic spectrum correlation, under the condition that SNR is more than or equal to 0dB, the identification rate reaches more than 90%, but the complexity of extracting the cyclic spectrum characteristic function is high. The document "E.like, V.Chakravarty, Zhijiang Wu.Reliable Modulation Classification at Low SNRUsing Spectral correction [ C ].4th IEEE CCNC. Las Vegas: IEEE Press,2007:1134 and 1138" uses the distribution diagram of the cyclic frequency alpha and the frequency f of the cyclic Spectral coefficient as the characteristic parameter, and can still realize higher recognition rate under the condition of lower SNR, but needs more input parameters.
Disclosure of Invention
The invention provides a deep learning intelligent modulation identification method based on cyclic spectrum estimation, aiming at overcoming the technical defects of high complexity and more input parameters of the existing characteristic parameter extraction identification method.
In order to solve the technical problems, the technical scheme of the invention is as follows:
the deep learning intelligent modulation identification method based on the cyclic spectrum estimation comprises the following steps:
s1: generating a modulated signal according to the carrier frequency and the symbol rate;
s2: carrying out cyclic spectrum estimation on the modulation signal, and extracting a cross-sectional diagram of a cyclic spectrum function;
s3: taking the sectional drawing as a characteristic, and training a deep neural network;
s4: and identifying the modulation mode of the unknown signal by using the deep neural network.
Wherein, the step S1 specifically includes: at the frequency fs16000Hz, carrier frequency fc5000Hz, symbol rate fdUnder the condition of 1200Hz, four modulation signals of 2ASK, 2FSK, BPSK and QPSK are generated and propagated through a gaussian channel.
In step S2, the calculation formula of the cyclic spectrum estimation specifically includes:
wherein,<·>trepresents time-averaged statistics;a cross-sectional view representing a cyclic spectral function.
Wherein, a time smoothing method is adopted forAnd estimating, wherein the specific calculation formula is as follows:
wherein w [ n ] represents a data sliding window; n' is the short-time FFT size; n is the observation time length.
Wherein, the step S3 specifically includes the following steps:
s31: cross-sectional view of cyclic spectrum functionAs characteristic parameters, input into the deep neural network from the input layerPerforming the following steps;
s32: normalizing the input value through the hidden layer, stabilizing the input value and reducing the influence of the front layer parameters on the rear layer parameters;
s33: and the output layer adopts one hot coding and outputs corresponding to different signal modulation modes.
Wherein, the step S4 specifically includes: and inputting the unknown signal into the deep neural network, and identifying to obtain the modulation mode of the unknown signal.
Compared with the prior art, the technical scheme of the invention has the beneficial effects that:
according to the deep learning intelligent modulation recognition method based on the cyclic spectrum estimation, the cyclic spectrum estimation and the deep neural network are combined, and the performance of the whole signal modulation recognition system is improved by utilizing the intelligent processing capability of the neural network and the better classification recognition capability of the cyclic spectrum; only the cross-sectional diagram of the cyclic spectrum function is utilized, the step of extracting cyclic spectrum features is omitted, and the time complexity of the algorithm is reduced.
Drawings
FIG. 1 is a schematic flow diagram of the process;
FIG. 2 is a schematic diagram of a deep neural network;
FIG. 3 is a cross-sectional view of cyclic spectrum functions of four modulation signals of 2ASK, 2FSK, BPSK and QPSK;
fig. 4 is a graph of the identification rates of four modulation signals, 2ASK, 2FSK, BPSK and QPSK.
Detailed Description
The drawings are for illustrative purposes only and are not to be construed as limiting the patent;
for the purpose of better illustrating the embodiments, certain features of the drawings may be omitted, enlarged or reduced, and do not represent the size of an actual product;
it will be understood by those skilled in the art that certain well-known structures in the drawings and descriptions thereof may be omitted.
The technical solution of the present invention is further described below with reference to the accompanying drawings and examples.
Example 1
As shown in fig. 1, the deep learning intelligent modulation identification method based on cyclic spectrum estimation includes the following steps:
s1: generating a modulated signal according to the carrier frequency and the symbol rate;
s2: carrying out cyclic spectrum estimation on the modulation signal, and extracting a cross-sectional diagram of a cyclic spectrum function;
s3: taking the sectional drawing as a characteristic, and training a deep neural network;
s4: and identifying the modulation mode of the unknown signal by using the deep neural network.
More specifically, the step S1 specifically includes: at the frequency fs16000Hz, carrier frequency fc5000Hz, symbol rate fdUnder the condition of 1200Hz, four modulation signals of 2ASK, 2FSK, BPSK and QPSK are generated and propagated through a gaussian channel.
In step S2, the calculation formula of the cyclic spectrum estimation specifically includes:
wherein,<·>trepresents time-averaged statistics;a cross-sectional view representing a cyclic spectral function.
More specifically, the time smoothing method is adopted forAnd estimating, wherein the specific calculation formula is as follows:
wherein w [ n ] represents a data sliding window; n' is the short-time FFT size; n is the observation time length.
More specifically, as shown in fig. 2, the step S3 specifically includes the following steps:
s31: cross-sectional view of cyclic spectrum functionAs characteristic parameters, inputting the characteristic parameters into the deep neural network by an input layer;
s32: normalizing the input value through the hidden layer, stabilizing the input value and reducing the influence of the front layer parameters on the rear layer parameters;
s33: and the output layer adopts one hot coding and outputs corresponding to different signal modulation modes.
More specifically, the step S4 specifically includes: and inputting the unknown signal into the deep neural network, and identifying to obtain the modulation mode of the unknown signal.
Example 2
More specifically, on the basis of embodiment 1, the method of the present invention is further described by taking the example of identifying four modulation signals of 2ASK, 2FSK, BPSK, and QPSK.
In practice, at a frequency fs16000Hz, carrier frequency fc5000Hz, symbol rate fdUnder the condition of 1200Hz, four modulation signals of 2ASK, 2FSK, BPSK and QPSK are generated, and the four modulation signals are propagated through a Gaussian channel; calculating the cyclic spectrum of the four signals, and extracting the section diagram of the cyclic spectrum functionAnd printing corresponding labels as original training data.
In the specific implementation process, the calculation formula of the cyclic spectrum estimation specifically is as follows:
wherein,<·>trepresents time-averaged statistics;a cross-sectional view representing a cyclic spectral function.
More specifically, the time smoothing method is adopted forAnd estimating, wherein the specific calculation formula is as follows:
wherein, w [ n ]]Representing a sliding window of data; n' is the short-time FFT size; n is the observation time length. According to the symmetry of the cyclic spectrum, the invention extracts the cross-sectional diagram of the cyclic spectrum functionAs a characteristic parameter, a cross-sectional view of the cyclic spectrum α of the four signals is shown in fig. 3.
In a specific implementation process, a deep neural network is trained, wherein the deep neural network comprises an input layer, three hidden layers and an output layer. The input parameters of the input layer are cross-sectional diagrams of the extracted cyclic spectrum function of the signalThe hidden layers of the front two layers use Bacth _ normalization to normalize the input value, so that the input value is more stable, the influence of parameters of the front layer on the rear layer is reduced, and the output layer adopts one hot coding and corresponds to different signal modulation modes.
In the specific implementation process, a signal with unknown modulation mode is generated and is receivedCross section diagram of signal cyclic spectrum function of unknown modulation modeInputting the signal into a deep neural network model, and predicting the modulation mode of the signal. Finally, the recognition rate curves of the signals of four different modulation modes between SNR-10 dB and SNR-10 dB are obtained, as shown in figure 4.
More specifically, the comparison of the computational complexity based on different modulation identification methods of the cyclic spectrum is shown in table 1:
TABLE 1 comparison table of computation time complexity of different modulation recognition algorithms based on cyclic spectrum
In the specific implementation process, the modulation identification algorithm based on the cyclic spectrum characteristic parameter extraction in table 1 refers to an algorithm adopted in a spectral correlation method [ J ] of modulation analysis and identification, 2001(03):34-36+46 "; the Modulation recognition algorithm based on the cyclic spectrum alpha and f sectional diagrams refers to the algorithm adopted in the literature, "E.like, V.Chakravarthy, Zhijiang Wu.Reliable Modulation Classification at Low SNRUsing Spectral correction [ C ].4th IEEE CCNC. Las Vegas: IEEE Press,2007: 1134-; a modulation identification algorithm based on a cyclic spectrum alpha sectional diagram is adopted as the method; according to the literature, "han dong, chu bin, wu jiangxing", spectral correlation method for modulation analysis and identification [ J ]. systematic engineering and electronic technology, 2001(03):34-36+46 ", the algorithm achieves an identification rate of more than 90% when SNR is 0 dB; according to the document "E.like, V.Chakravarty, Zhijiang Wu.Reliable Modulation Classification at Low SNRUsing Spectral correction [ C ].4th IEEE CCNC. Las Vegas: IEEE Press,2007: 1134-. According to the results shown in fig. 4, the algorithm adopted by the invention achieves a recognition rate of 97% when the SNR is-5 dB, and still has good recognition performance at a lower SNR.
In the concrete implementation process, the inventionThe deep learning intelligent modulation recognition method based on the cyclic spectrum estimation combines the cyclic spectrum estimation and the deep neural network, and improves the performance of the whole signal modulation recognition system by utilizing the intelligent processing capability of the neural network and the better classification recognition capability of the cyclic spectrum; meanwhile, compared with the existing modulation identification scheme based on cyclic spectrum characteristic parameter extraction, the method only uses the section diagram of the cyclic spectrum functionThe step of extracting the cyclic spectrum features is omitted, so that the time complexity of the algorithm is O (n)2) Reduced to O (n), compared with the existing scheme, the invention adopts an innovative mixed structure, and realizes better identification performance with lower complexity.
It should be understood that the above-described embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not intended to limit the embodiments of the present invention. Other variations and modifications will be apparent to persons skilled in the art in light of the above description. And are neither required nor exhaustive of all embodiments. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present invention should be included in the protection scope of the claims of the present invention.
Claims (6)
1. The deep learning intelligent modulation identification method based on the cyclic spectrum estimation is characterized by comprising the following steps of:
s1: generating a modulated signal according to the carrier frequency and the symbol rate;
s2: carrying out cyclic spectrum estimation on the modulation signal, and extracting a cross-sectional diagram of a cyclic spectrum function;
s3: taking the sectional drawing as a characteristic, and training a deep neural network;
s4: and identifying the modulation mode of the unknown signal by using the deep neural network.
2. The cyclic spectrum estimation-based deep learning intelligent modulation identification method according to claim 1, whichIs characterized in that: the step S1 specifically includes: at the frequency fs16000Hz, carrier frequency fc5000Hz, symbol rate fdUnder the condition of 1200Hz, four modulation signals of 2ASK, 2FSK, BPSK and QPSK are generated and propagated through a gaussian channel.
3. The cycle spectrum estimation-based deep learning intelligent modulation identification method according to claim 2, characterized in that: in step S2, the calculation formula of the cyclic spectrum estimation specifically includes:
wherein,<·>trepresents time-averaged statistics;a cross-sectional view representing a cyclic spectral function.
4. The cycle spectrum estimation-based deep learning intelligent modulation identification method according to claim 3, characterized in that: using a time smoothing method toAnd estimating, wherein the specific calculation formula is as follows:
wherein w [ n ] represents a data sliding window; n' is the short-time FFT size; n is the observation time length.
5. The cycle spectrum estimation-based deep learning intelligent modulation identification method according to claim 3, characterized in that: the step S3 specifically includes the following steps:
s31: cross-sectional view of cyclic spectrum functionAs characteristic parameters, inputting the characteristic parameters into the deep neural network by an input layer;
s32: normalizing the input value through the hidden layer, stabilizing the input value and reducing the influence of the front layer parameters on the rear layer parameters;
s33: and the output layer adopts one hot coding and outputs corresponding to different signal modulation modes.
6. The cycle spectrum estimation-based deep learning intelligent modulation identification method according to claim 4, characterized in that: the step S4 specifically includes: and inputting the unknown signal into the deep neural network, and identifying to obtain the modulation mode of the unknown signal.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910594826.8A CN110321953A (en) | 2019-07-03 | 2019-07-03 | Deep learning intelligent modulation recognition methods based on circulation Power estimation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910594826.8A CN110321953A (en) | 2019-07-03 | 2019-07-03 | Deep learning intelligent modulation recognition methods based on circulation Power estimation |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110321953A true CN110321953A (en) | 2019-10-11 |
Family
ID=68122431
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910594826.8A Pending CN110321953A (en) | 2019-07-03 | 2019-07-03 | Deep learning intelligent modulation recognition methods based on circulation Power estimation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110321953A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111507205A (en) * | 2020-03-27 | 2020-08-07 | 中国人民解放军海军工程大学 | Modulation recognition device based on signal cyclic spectrum and deep learning and use method thereof |
| CN111639595A (en) * | 2020-05-29 | 2020-09-08 | 桂林电子科技大学 | Unmanned aerial vehicle micro-motion characteristic signal detection method based on weight-agnostic neural network |
| CN111800357A (en) * | 2020-07-03 | 2020-10-20 | 全球能源互联网研究院有限公司 | Method and system for distinguishing modulation types based on cyclic spectrum |
| CN112600775A (en) * | 2020-11-18 | 2021-04-02 | 江苏师范大学 | Modulation mode identification method based on cyclic correlation entropy spectrum peak position |
| WO2021082469A1 (en) * | 2019-10-28 | 2021-05-06 | 北京邮电大学 | Modulation mode identification method and apparatus |
| CN114024808A (en) * | 2021-09-22 | 2022-02-08 | 中山大学 | Modulation signal identification method and system based on deep learning |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106130942A (en) * | 2016-07-05 | 2016-11-16 | 东南大学 | A kind of wireless communication signals Modulation Identification based on Cyclic Spectrum and method for parameter estimation |
| CN108566253A (en) * | 2018-02-12 | 2018-09-21 | 北京邮电大学 | It is a kind of based on the signal recognition method extracted to power spectrum signal fit characteristic |
| CN108600135A (en) * | 2018-04-27 | 2018-09-28 | 中国科学院计算技术研究所 | A kind of recognition methods of signal modulation mode |
| CN109117747A (en) * | 2018-07-23 | 2019-01-01 | 电子科技大学 | Radar signal classification method based on cyclo-stationary Yu depth convolutional neural networks |
| CN109802905A (en) * | 2018-12-27 | 2019-05-24 | 西安电子科技大学 | Digital signal Automatic Modulation Recognition method based on CNN convolutional neural networks |
| CN109818892A (en) * | 2019-01-18 | 2019-05-28 | 华中科技大学 | Construct Cyclic Spectrum characteristic parameter extraction model and signal modulation mode recognition methods |
-
2019
- 2019-07-03 CN CN201910594826.8A patent/CN110321953A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106130942A (en) * | 2016-07-05 | 2016-11-16 | 东南大学 | A kind of wireless communication signals Modulation Identification based on Cyclic Spectrum and method for parameter estimation |
| CN108566253A (en) * | 2018-02-12 | 2018-09-21 | 北京邮电大学 | It is a kind of based on the signal recognition method extracted to power spectrum signal fit characteristic |
| CN108600135A (en) * | 2018-04-27 | 2018-09-28 | 中国科学院计算技术研究所 | A kind of recognition methods of signal modulation mode |
| CN109117747A (en) * | 2018-07-23 | 2019-01-01 | 电子科技大学 | Radar signal classification method based on cyclo-stationary Yu depth convolutional neural networks |
| CN109802905A (en) * | 2018-12-27 | 2019-05-24 | 西安电子科技大学 | Digital signal Automatic Modulation Recognition method based on CNN convolutional neural networks |
| CN109818892A (en) * | 2019-01-18 | 2019-05-28 | 华中科技大学 | Construct Cyclic Spectrum characteristic parameter extraction model and signal modulation mode recognition methods |
Non-Patent Citations (4)
| Title |
|---|
| 刘鹭航 等: "基于循环谱的相位编码信号调制参数估计", 《遥测遥控》 * |
| 张春杰等: "PSK信号的调制参数估计与识别", 《应用科技》 * |
| 管鹏辉 等: "一种基于循环谱分析的调制信号识别算法", 《数据通信》 * |
| 陈雪等: "一种优化的卷积神经网络调制识别算法", 《电讯技术》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021082469A1 (en) * | 2019-10-28 | 2021-05-06 | 北京邮电大学 | Modulation mode identification method and apparatus |
| CN111507205A (en) * | 2020-03-27 | 2020-08-07 | 中国人民解放军海军工程大学 | Modulation recognition device based on signal cyclic spectrum and deep learning and use method thereof |
| CN111639595A (en) * | 2020-05-29 | 2020-09-08 | 桂林电子科技大学 | Unmanned aerial vehicle micro-motion characteristic signal detection method based on weight-agnostic neural network |
| CN111639595B (en) * | 2020-05-29 | 2022-03-18 | 桂林电子科技大学 | Unmanned aerial vehicle micro-motion characteristic signal detection method based on weight-agnostic neural network |
| CN111800357A (en) * | 2020-07-03 | 2020-10-20 | 全球能源互联网研究院有限公司 | Method and system for distinguishing modulation types based on cyclic spectrum |
| CN112600775A (en) * | 2020-11-18 | 2021-04-02 | 江苏师范大学 | Modulation mode identification method based on cyclic correlation entropy spectrum peak position |
| CN114024808A (en) * | 2021-09-22 | 2022-02-08 | 中山大学 | Modulation signal identification method and system based on deep learning |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110321953A (en) | Deep learning intelligent modulation recognition methods based on circulation Power estimation | |
| CN104468001B (en) | Signal identification method and system based on radio signal frequency spectrum feature template | |
| CN108718288A (en) | Recognition of digital modulation schemes method based on convolutional neural networks | |
| CN106130942A (en) | A kind of wireless communication signals Modulation Identification based on Cyclic Spectrum and method for parameter estimation | |
| CN111444805B (en) | Improved multi-scale wavelet entropy digital signal modulation identification method | |
| CN112787964B (en) | BPSK and QPSK signal modulation identification method based on range median domain features | |
| CN107770108A (en) | A kind of combined modulation recognition methods of K mean clusters and classification training SVM classifier | |
| CN114422311B (en) | Signal modulation recognition method and system combining deep neural network and expert priori features | |
| Zhang et al. | Open set recognition of communication signal modulation based on deep learning | |
| CN114567528A (en) | Communication signal modulation mode open set identification method and system based on deep learning | |
| CN114024808A (en) | Modulation signal identification method and system based on deep learning | |
| Samala et al. | Spectrum Sensing Techniques in Cognitive Radio Technology: A Review Paper. | |
| Onumanyi et al. | Amplitude quantization method for autonomous threshold estimation in self-reconfigurable cognitive radio systems | |
| Liu et al. | A novel wireless interference identification and scheduling method based on convolutional neural network | |
| CN104467995B (en) | Blind primary user detection based on HMM and transmission level recognition methods | |
| CN115866615B (en) | Wireless network communication relation discovery method based on electromagnetic spectrum characteristics | |
| CN115913850B (en) | Open set modulation identification method based on residual error network | |
| CN116132237B (en) | OFDM signal detection and identification method based on periodic convolution and fractal | |
| CN103973381B (en) | Based on the cooperative spectrum detection method of Cholesky matrix decomposition | |
| Gai et al. | Spectrum sensing method based on residual cellular network | |
| An et al. | Spectrum sensing based on KNN algorithm for 230 MHz power private networks | |
| CN111404852B (en) | Modulation mode identification method based on amplitude and spectral amplitude characteristics | |
| CN111988252A (en) | Signal modulation mode identification method based on deep learning | |
| Xu et al. | PCA-RNN-based intelligent mobile drone spectrum sensing algorithm | |
| Tekbıyık et al. | Real-world considerations for deep learning in wireless signal identification based on spectral correlation function |
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 |
Application publication date: 20191011 |
|
| RJ01 | Rejection of invention patent application after publication |