CN113315727B - Digital communication signal modulation identification method based on preprocessing noise reduction - Google Patents
Digital communication signal modulation identification method based on preprocessing noise reduction Download PDFInfo
- Publication number
- CN113315727B CN113315727B CN202110333486.0A CN202110333486A CN113315727B CN 113315727 B CN113315727 B CN 113315727B CN 202110333486 A CN202110333486 A CN 202110333486A CN 113315727 B CN113315727 B CN 113315727B
- Authority
- CN
- China
- Prior art keywords
- signal
- noise
- wavelet
- modulation
- noise reduction
- 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
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
- G06F18/24—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
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield 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
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/30—Noise filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/52—Scale-space analysis, e.g. wavelet analysis
-
- 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
- G06F2218/04—Denoising
- G06F2218/06—Denoising by applying a scale-space analysis, e.g. using wavelet analysis
-
- 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
-
- 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)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Molecular Biology (AREA)
- Multimedia (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Noise Elimination (AREA)
Abstract
The invention provides a digital communication signal modulation identification method based on preprocessing noise reduction, which comprises the following steps: s1: constructing a modulation signal according to the carrier frequency, the code element rate and the sampling frequency, and carrying out noise processing; s2: carrying out noise reduction pretreatment on the signal subjected to noise addition treatment by using a self-adaptive filtering technology; s3: extracting wavelet transformation characteristics and high-order accumulation characteristics of the signals subjected to noise reduction preprocessing; s4: inputting the wavelet transformation characteristics and the high-order accumulation characteristics into a BP neural network for network training; s5: and identifying the modulation mode of the unknown signal by using the trained BP neural network. The invention uses the self-adaptive filtering technology to preprocess the communication signal containing noise so as to improve the signal-to-noise ratio of the signal, then extracts the wavelet transformation and the high-order cumulant characteristic of the signal, and finally inputs the signal into the neural network classifier, thereby obtaining better recognition effect under the environment of low signal-to-noise ratio.
Description
Technical Field
The invention relates to the technical field of communication, in particular to a digital communication signal modulation identification method based on preprocessing noise reduction.
Background
Along with the increasing complexity of communication signal systems and modulation patterns, the signal environment is increasingly severe, and the identification of modulation types in non-cooperative communication becomes important in the military field and the civil field. The signal-to-noise ratio of a signal in actual communication may reach a very low value, and at present, the automatic modulation and identification method of a communication signal under a low signal-to-noise ratio has a poor effect.
In 1986, hipp et al proposed for the first time a method for identifying modulation by using high-order cumulant of signals as features, and since more developments were made on the research on the high-order cumulant feature parameters, the method is widely applied to the modulation identification algorithm of signals. Because the high-order cumulant characteristic of the signal has good noise immunity because the high-order cumulant of gaussian white noise is zero.
In 1995 to 2000, ho et al proposed a method for extracting characteristic parameters by using Haar wavelet transform of signals to perform modulation identification, and completed classification of MPSK and MFSK signals. The wavelet transform feature extraction method can express the feature information of the signals through transformation, and is convenient to observe and analyze. But the scale selection for wavelet transformation has no fixed rule, and the problem of low multi-path fading channel characteristic extraction discrimination cannot be solved.
Adaptive filtering is an optimal filtering method developed in recent years. The method is an optimal filtering method developed on the basis of linear filtering such as wiener filtering, kalman filtering and the like. Because it has stronger adaptability and better filtering performance. Therefore, the method is widely applied to engineering practice, particularly information processing technology. Adaptive filtering exists in many different fields such as signal processing, control, image processing, etc., and is an intelligent and more targeted filtering method, which is generally used for denoising.
Published as 2014, 05 and 07, and published as CN103780462A, discloses a satellite communication signal modulation identification method based on high-order cumulant and spectral characteristics, which comprises the following steps: the method comprises the steps of signal band-pass filtering, carrier frequency estimation, symbol rate estimation, high-order cumulant parameter acquisition, APSK or 16QAM signal identification, quadratic spectrum peak number acquisition, BPSK or MSK signal identification, quadratic spectrum acquisition, quadratic spectrum peak number acquisition, pi/4 DQPSK signal identification, baseband quadratic spectrum peak number acquisition, 6PSK, 8PSK, OQPSK and QPSK signal identification. The patent does not recognize the modulation mode of low signal-to-noise ratio signals in actual communication.
Disclosure of Invention
The invention provides a digital communication signal modulation identification method based on preprocessing noise reduction, which can obtain a better identification effect in a low signal-to-noise ratio environment.
In order to solve the technical problems, the technical scheme of the invention is as follows:
a digital communication signal modulation identification method based on preprocessing noise reduction comprises the following steps:
s1: constructing a modulation signal according to the carrier frequency, the code element rate and the sampling frequency, and performing noise processing;
s2: carrying out noise reduction pretreatment on the signal subjected to noise addition treatment by using a self-adaptive filtering technology;
s3: extracting wavelet transformation characteristics and high-order accumulation characteristics of the signals subjected to noise reduction preprocessing;
s4: inputting the wavelet transformation characteristics and the high-order accumulation characteristics into a BP neural network for network training;
s5: and identifying the modulation mode of the unknown signal by using the trained BP neural network.
Preferably, in step S1, six modulation signals, i.e., 2ASK, 4ASK, BPSK, QPSK, 2FSK, and 4FSK, are constructed according to the carrier frequency, symbol rate, and sampling frequency, and are denoted as S (n).
Preferably, the noise adding process in step S1 specifically includes:
six kinds of modulation signals are propagated through a Gaussian white noise channel, and the noise is recorded as v 0 (n) the signal after noise addition is d (n) = s (n) + v 0 (n)
Preferably, the adaptive filtering technique in step S2 is specifically an adaptive noise canceller, two inputs of the adaptive noise canceller are an original input and a reference input, respectively, where the original input is a signal d (n) after adding noise, the reference input is a signal y (n) after filtering an actually acquired noise signal v (n), and an output signal of the adaptive noise canceller is an error signal e (n) = d (n) -y (n) = S (n) + v 0 (n)-y(n)。
Preferably, the wavelet transform characteristic in step S3 is a wavelet transform coefficient magnitude.
Preferably, the wavelet transform coefficient amplitude is calculated as follows:
the formula of the wavelet transform is:
wherein, a is a scale factor,b is a translation factor, denotes the complex conjugate, Ψ (t) is a wavelet function, Ψ (a.b) (t) is a wavelet basis function obtained by performing telescopic translation on a wavelet mother function:
the method comprises the following steps of selecting a Haar wavelet to perform wavelet transformation on a received signal, wherein a Haar wavelet function and a wavelet basis function thereof are defined as follows:
the wavelet transform of various types of modulation signals in the same symbol can be expressed as:
modulo the result of the wavelet transform can result in:
preferably, the wavelet transform coefficient amplitudes of the obtained various signals are filtered through a median filter to obtain wavelet transform characteristics.
Preferably, the high-order cumulant feature specifically is:
the formula of the high-order cumulant is specifically as follows:
for a zero-mean stationary random process { x (t) }, sampling the random process at any k moments, and defining the k-order cumulant as:
C kx (τ 1 ,τ 2 ,...,τ k-1 )=Cum(x(t),x(t+τ 1 ),...,x(t+τ k-1 ))
wherein, cum (-) means to calculate the cumulant of (-) and the p-order q-degree mixing moment is:
M pq =E{[x(t) p-q x * (t) q ]}
where denotes the conjugation of the signal, q denotes the number of conjugates, and the meaning of E (-) is the mathematical expectation of the (-) equation;
the expression of the cumulative quantity of each order of the stationary random process { x (t) } which can obtain the zero mean value is as follows:
C 20 =Cum(x,x)=M 20 ,C 21 =Cum(x,x * )=M 21 ,
wherein denotes a complex conjugate;wherein p represents the order of the higher order cumulant and q represents the conjugation position; and selecting | C40|, and the ratio of | C60| to | C40|, as the characteristics of the high-order accumulated quantity.
Preferably, the BP neural network in step S4 includes an input layer, a single hidden layer, and an output layer, where the input parameters of the input layer are the wavelet transformation characteristic and the high-order cumulant characteristic, and the output layer outputs a modulation mode of a signal.
Preferably, the number of nodes of the single hidden layer of the BP neural network is 8.
Compared with the prior art, the technical scheme of the invention has the beneficial effects that:
the invention uses the self-adaptive filtering technology to preprocess the communication signal containing noise so as to improve the signal-to-noise ratio of the signal, then extracts the wavelet transformation and high-order cumulant characteristics of the signal, finally inputs the wavelet transformation and high-order cumulant characteristics into the neural network classifier, utilizes the good anti-noise performance of the wavelet transformation and high-order cumulant and the better classification and identification capability of the BP neural network, improves the performance of the whole digital communication signal automatic modulation and identification system, can obtain better identification effect in the environment with low signal-to-noise ratio and has higher identification rate.
Drawings
FIG. 1 is a schematic flow chart of the method of the present invention.
Fig. 2 is a schematic diagram of the variance of the wavelet transform amplitude of the signals of six modulation modes in the embodiment.
Fig. 3 is a diagram of | C40| of the high-order accumulated amount of signals of six modulation modes in the embodiment.
Fig. 4 is a graph illustrating the identification rate curves of the six modulation mode signals in the embodiment between SNR = -4dB and 10 dB.
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 with reference to the drawings and the embodiments.
Example 1
The embodiment provides a digital communication signal modulation identification method based on preprocessing noise reduction, as shown in fig. 1, including the following steps:
s1: constructing a modulation signal according to the carrier frequency, the code element rate and the sampling frequency, and performing noise processing;
s2: carrying out noise reduction pretreatment on the signal subjected to noise addition treatment by using a self-adaptive filtering technology;
s3: extracting wavelet transformation characteristics and high-order accumulation characteristics of the signals subjected to denoising pretreatment;
s4: inputting the wavelet transformation characteristics and the high-order accumulation characteristics into a BP neural network for network training;
s5: and identifying the modulation mode of the unknown signal by using the trained BP neural network.
In this embodiment, in step S1, under the conditions that the sampling frequency fs =10kHz, the carrier frequency fc =1kHz, and the symbol rate fd =100Hz, six modulation signals, i.e., 2ASK, 4ASK, BPSK, QPSK, 2FSK, and 4FSK, are generated, and are denoted as S (n).
The noise adding process in the step S1 specifically comprises the following steps:
six kinds of modulation signals are propagated through a Gaussian white noise channel, and the noise is recorded as v 0 (n) the signal after noise addition is d (n) = s (n) + v 0 (n)
In step S2, the adaptive filtering technique is specifically an adaptive noise canceller, two inputs of the adaptive noise canceller are an original input and a reference input, respectively, where the original input is a signal d (n) after adding noise, the reference input is a signal y (n) after filtering a noise signal v (n) actually acquired, and an output signal of the adaptive noise canceller is an error signal e (n) = d (n) -y (n) = S (n) + v 0 (n)-y(n)。
The wavelet transform characteristic in step S3 is a wavelet transform coefficient magnitude.
The wavelet transform coefficient amplitude is calculated as follows:
the formula of the wavelet transform is:
where a is a scale factor, b is a translation factor, denotes the complex conjugate, Ψ (t) is a wavelet function, Ψ (a.b) (t) is a wavelet basis function obtained by performing telescopic translation on a wavelet mother function:
the method comprises the following steps of selecting a Haar wavelet to perform wavelet transformation on a received signal, wherein a Haar wavelet function and a wavelet basis function thereof are defined as follows:
the wavelet transform of various types of modulation signals in the same symbol can be expressed as:
taking the modulus of the result of the wavelet transform can obtain:
the wavelet transform coefficient amplitudes of the obtained various signals need to be filtered through a median filter to remove peaks, the variance of the wavelet transform amplitudes of the six signals is shown in fig. 2 to obtain wavelet transform characteristics, and for the PSK signals, the variance σ of | CWT | is idealized to be zero. Conversely, for ASK and FSK modulation, the variance σ of | CWT | will be greater than zero. According to this feature, PSK signals can be distinguished from ASK, FSK signals.
The high-order cumulant features specifically include:
the formula of the high-order cumulant is specifically as follows:
for a zero-mean stationary random process { x (t) }, sampling the random process at any k time, the k-order cumulant is defined as:
C kx (τ 1 ,τ 2 ,...,τ k-1 )=Cum(x(t),x(t+τ 1 ),...,x(t+τ k-1 ))
where, cum (-) means to sum up (·), the p-order q-degree mixing moment is:
M pq =E{[x(t) p-q x * (t) q ]}
where denotes the conjugation of the signal, q denotes the number of conjugates, and the meaning of E (-) is the mathematical expectation of the (-) equation;
the expression of the cumulative quantity of each order of the stationary random process { x (t) } which can obtain the zero mean value is as follows:
C 20 =Cum(x,x)=M 20 ,C 21 =Cum(x,x * )=M 21 ,
wherein, denotes a complex conjugate;wherein p represents the order of the higher order cumulant and q represents the conjugation position; selecting | C40| and a ratio of | C60| to | C40| as characteristics of high-order cumulant, wherein when the phase of a received signal changes, the calculation result of the high-order cumulant may appear alternately in positive and negative directions, and directly taking absolute values of the high-order cumulant according to the unstable influence factor;for the influence of the signal amplitude, a ratio form is adopted when the characteristic parameters are selected. The two selected characteristics are | C40| and | C60|/| C40| respectively. The | C40| of the high-order accumulated amounts of the six kinds of signals is shown in fig. 3.
The BP neural network of the step S4 comprises an input layer, a single hidden layer and an output layer, wherein input parameters of the input layer are the wavelet transformation characteristics and the high-order cumulant characteristics, and the output layer outputs a modulation mode of signals.
The number of nodes of the single hidden layer of the BP neural network is 8.
According to the graph shown in fig. 4, the algorithm adopted by the invention achieves the recognition rate of 99.5% when the SNR is = -4dB, and still has good recognition performance at lower signal to noise ratio.
Meanwhile, compared with the existing modulation identification method based on wavelet transformation and high-order cumulant characteristic parameter extraction, the method only uses three characteristic parameters, and realizes better identification performance with lower complexity
The same or similar reference numerals correspond to the same or similar parts;
the terms describing positional relationships in the drawings are for illustrative purposes only and should not be construed as limiting the patent;
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 (9)
1. A digital communication signal modulation identification method based on preprocessing noise reduction is characterized by comprising the following steps:
s1: constructing a modulation signal according to the carrier frequency, the code element rate and the sampling frequency, and performing noise processing;
s2: carrying out noise reduction pretreatment on the signal subjected to noise addition treatment by using a self-adaptive filtering technology;
s3: extracting wavelet transformation characteristics and high-order cumulant characteristics of the signals subjected to denoising pretreatment;
s4: inputting the wavelet transformation characteristics and the high-order cumulant characteristics into a BP neural network for network training;
s5: recognizing the modulation mode of the unknown signal by using the trained BP neural network;
the high-order cumulant characteristic specifically comprises the following steps:
the formula of the high-order cumulant is specifically as follows:
for a zero-mean stationary random process { x (t) }, sampling the random process at any k moments, and defining the k-order cumulant as:
C kx (τ 1 ,τ 2 ,...,τ k-1 )=Cum(x(t),x(t+τ 1 ),...,x(t+τ k-1 ))
wherein, cum (-) means to calculate the cumulant of (-) and the p-order q-degree mixing moment is:
M pq =E{[x(t) p-q x * (t) q ]}
where denotes the conjugation of the signal, q denotes the number of conjugates, and the meaning of E (-) is the mathematical expectation of the (-) equation;
the expression of the cumulative quantity of each order of the smooth random process { x (t) } which can obtain the zero mean value is as follows:
C 20 =Cum(x,x)=M 20 ,C 21 =Cum(x,x * )=M 21 ,
wherein denotes a complex conjugate; c mn Where m represents the order of the high order cumulant and n represents the conjugation position; selecting | C 40 L, and | C 60 I and I C 40 The ratio of | is characterized as a high-order cumulant.
2. The method according to claim 1, wherein six modulation signals, i.e. 2ASK, 4ASK, BPSK, QPSK, 2FSK and 4FSK, are constructed in step S1 according to carrier frequency, symbol rate and sampling frequency, and are denoted as S (n).
3. The method for modulating and identifying the digital communication signal based on the preprocessing noise reduction according to claim 2, wherein the noise adding process in the step S1 is specifically as follows:
six kinds of modulation signals are propagated through a Gaussian white noise channel, and the noise is recorded as v 0 (n) the signal after noise addition is d (n) = s (n) + v 0 (n)。
4. The method according to claim 3, wherein the adaptive filtering technique in step S2 is specifically an adaptive noise canceller, and two inputs of the adaptive noise canceller are an original input and a reference input, respectively, wherein the original input is a signal d (n) after adding noise, the reference input is a signal y (n) after filtering an actually acquired noise signal v (n), and an output signal of the adaptive noise canceller is an error signal e (n) = nd(n)-y(n)=s(n)+v 0 (n)-y(n)。
5. The method for modulating and identifying digital communication signals based on preprocessing noise reduction as claimed in claim 4, wherein the wavelet transform characteristic in step S3 is wavelet transform coefficient amplitude.
6. The method for modulating and identifying a digital communication signal based on preprocessing noise reduction according to claim 5, wherein the amplitude of the wavelet transform coefficients is calculated as follows:
the formula of the wavelet transform is:
where a is a scale factor, b is a translation factor, denotes the complex conjugate, Ψ (t) is a wavelet function, Ψ (a.b) (t) is a wavelet basis function obtained by performing telescopic translation on a wavelet mother function:
the method comprises the following steps of selecting a Haar wavelet to perform wavelet transformation on a received signal, wherein a Haar wavelet function and a wavelet basis function thereof are defined as follows:
the wavelet transform of various types of modulation signals in the same symbol can be expressed as:
taking the modulus of the result of the wavelet transform can obtain:
7. the method of claim 6, wherein the wavelet transform coefficients of each type of signal obtained are further filtered by a median filter to obtain wavelet transform characteristics.
8. The preprocessing noise reduction-based digital communication signal modulation and recognition method according to claim 7, wherein the BP neural network of the step S4 comprises an input layer, a single hidden layer and an output layer, wherein input parameters of the input layer are the wavelet transformation characteristics and the high-order cumulant characteristics, and the output layer outputs a modulation mode of signals.
9. The preprocessing noise reduction-based digital communication signal modulation recognition method of claim 8, wherein the number of nodes of the single hidden layer of the BP neural network is 8.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110333486.0A CN113315727B (en) | 2021-03-29 | 2021-03-29 | Digital communication signal modulation identification method based on preprocessing noise reduction |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110333486.0A CN113315727B (en) | 2021-03-29 | 2021-03-29 | Digital communication signal modulation identification method based on preprocessing noise reduction |
Publications (2)
Publication Number | Publication Date |
---|---|
CN113315727A CN113315727A (en) | 2021-08-27 |
CN113315727B true CN113315727B (en) | 2022-10-11 |
Family
ID=77371942
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110333486.0A Active CN113315727B (en) | 2021-03-29 | 2021-03-29 | Digital communication signal modulation identification method based on preprocessing noise reduction |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113315727B (en) |
Families Citing this family (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114422016B (en) * | 2021-12-28 | 2024-01-02 | 成都爱科特科技发展有限公司 | PCMA signal identification method |
CN114398931A (en) * | 2022-01-17 | 2022-04-26 | 中国人民解放军国防科技大学 | Modulation recognition method and system based on numerical characteristic and image characteristic fusion |
CN114826850B (en) * | 2022-04-21 | 2022-11-25 | 中国人民解放军国防科技大学 | Modulation identification method, device and equipment based on time-frequency diagram and deep learning |
CN115225441A (en) * | 2022-07-21 | 2022-10-21 | 南京航空航天大学 | Unmanned aerial vehicle cluster communication waveform identification method in complex environment |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103067325A (en) * | 2013-01-31 | 2013-04-24 | 南京邮电大学 | Cooperative modulation identification method based on multi-class characteristic parameters and evidence theory |
CN108768907A (en) * | 2018-01-05 | 2018-11-06 | 南京邮电大学 | A kind of Modulation Identification method based on temporal characteristics statistic and BP neural network |
CN110276256A (en) * | 2019-05-17 | 2019-09-24 | 中国人民解放军战略支援部队信息工程大学 | Based on the low signal-to-noise ratio Modulation Recognition of Communication Signal method and device for adjusting ginseng accidental resonance |
CN110690931A (en) * | 2019-10-11 | 2020-01-14 | 中国科学院软件研究所 | Digital signal adaptive code rate estimation method and device based on multi-wavelet-base combination |
Family Cites Families (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8792564B2 (en) * | 2008-10-28 | 2014-07-29 | Sony Corporation | Adaptive preprocessing method using feature-extracted video maps |
US20200153535A1 (en) * | 2018-11-09 | 2020-05-14 | Bluecom Systems and Consulting LLC | Reinforcement learning based cognitive anti-jamming communications system and method |
CN111049770B (en) * | 2019-12-06 | 2021-11-02 | 西安电子科技大学 | Modulation signal identification method based on high-order cumulant |
-
2021
- 2021-03-29 CN CN202110333486.0A patent/CN113315727B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103067325A (en) * | 2013-01-31 | 2013-04-24 | 南京邮电大学 | Cooperative modulation identification method based on multi-class characteristic parameters and evidence theory |
CN108768907A (en) * | 2018-01-05 | 2018-11-06 | 南京邮电大学 | A kind of Modulation Identification method based on temporal characteristics statistic and BP neural network |
CN110276256A (en) * | 2019-05-17 | 2019-09-24 | 中国人民解放军战略支援部队信息工程大学 | Based on the low signal-to-noise ratio Modulation Recognition of Communication Signal method and device for adjusting ginseng accidental resonance |
CN110690931A (en) * | 2019-10-11 | 2020-01-14 | 中国科学院软件研究所 | Digital signal adaptive code rate estimation method and device based on multi-wavelet-base combination |
Also Published As
Publication number | Publication date |
---|---|
CN113315727A (en) | 2021-08-27 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN113315727B (en) | Digital communication signal modulation identification method based on preprocessing noise reduction | |
CN108718288B (en) | Method for recognizing digital signal modulation mode based on convolutional neural network | |
Zeng et al. | Spectrum analysis and convolutional neural network for automatic modulation recognition | |
CN108234370B (en) | Communication signal modulation mode identification method based on convolutional neural network | |
CN112511477A (en) | Hybrid satellite communication modulation identification method and system based on constellation diagram and deep learning | |
CN111444805B (en) | Improved multi-scale wavelet entropy digital signal modulation identification method | |
CN111814777B (en) | Modulation pattern recognition method based on characteristic quantity grading | |
CN111327554B (en) | Feature extraction method for digital modulation signal identification | |
CN110690931B (en) | Digital signal adaptive code rate estimation method and device based on multi-wavelet-base combination | |
CN114422311B (en) | Signal modulation recognition method and system combining deep neural network and expert priori features | |
Tian et al. | Modulation pattern recognition of communication signals based on fractional low-order Choi-Williams distribution and convolutional neural network in impulsive noise environment | |
CN112737992A (en) | Underwater sound signal modulation mode self-adaptive in-class identification method | |
CN109076038B (en) | Method for estimating parameters of a signal contained in a frequency band | |
Hong et al. | Modulation classification of BPSK and QPSK signals using a two element antenna array receiver | |
CN113537025A (en) | Electromagnetic modulation signal deep denoising method and system based on self-supervision learning | |
Abdelmutalab et al. | Automatic modulation classification using polynomial classifiers | |
Su et al. | Dual-use of modulation recognition techniques for digital communication signals | |
CN105763499A (en) | CPFSK modulation recognizing method based on signal instantaneous frequency moving average | |
CN113162879B (en) | Modulation signal identification method combining feature extraction | |
Xu et al. | PSK/QAM modulation recognition by convolutional neural network | |
CN111814703B (en) | HB-based signal joint feature extraction method under non-reconstruction condition | |
CN113361433A (en) | Modulation signal identification method based on neural network and application thereof | |
CN112865822A (en) | Audio interference resisting method for analog-digital mixed amplitude modulation broadcast receiver | |
CN111917674A (en) | Modulation identification method based on deep learning | |
CN116933066A (en) | Wireless signal single-channel blind source separation method and device based on deep neural network |
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 |