CN116431991B - Satellite communication radiation source individual identification method based on phase modulation signal feature fusion - Google Patents
Satellite communication radiation source individual identification method based on phase modulation signal feature fusion Download PDFInfo
- Publication number
- CN116431991B CN116431991B CN202310479126.0A CN202310479126A CN116431991B CN 116431991 B CN116431991 B CN 116431991B CN 202310479126 A CN202310479126 A CN 202310479126A CN 116431991 B CN116431991 B CN 116431991B
- Authority
- CN
- China
- Prior art keywords
- matrix
- mapping
- radiation source
- satellite communication
- feature
- 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
- 230000005855 radiation Effects 0.000 title claims abstract description 56
- 238000004891 communication Methods 0.000 title claims abstract description 55
- 238000000034 method Methods 0.000 title claims abstract description 35
- 230000004927 fusion Effects 0.000 title claims abstract description 20
- 238000013507 mapping Methods 0.000 claims abstract description 64
- 239000011159 matrix material Substances 0.000 claims abstract description 63
- 238000012549 training Methods 0.000 claims abstract description 22
- 238000000354 decomposition reaction Methods 0.000 claims abstract description 15
- 238000001228 spectrum Methods 0.000 claims abstract description 11
- 238000012545 processing Methods 0.000 claims abstract description 7
- 230000008030 elimination Effects 0.000 claims abstract description 6
- 238000003379 elimination reaction Methods 0.000 claims abstract description 6
- 238000012360 testing method Methods 0.000 claims description 10
- 230000001186 cumulative effect Effects 0.000 claims description 6
- 238000010219 correlation analysis Methods 0.000 claims description 3
- 238000007781 pre-processing Methods 0.000 claims description 3
- 238000000513 principal component analysis Methods 0.000 claims description 3
- 230000003595 spectral effect Effects 0.000 claims description 3
- 238000013519 translation Methods 0.000 claims description 2
- 239000013598 vector Substances 0.000 claims description 2
- 238000005516 engineering process Methods 0.000 abstract description 4
- 230000006870 function Effects 0.000 description 6
- 230000003321 amplification Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000003199 nucleic acid amplification method Methods 0.000 description 2
- 238000007500 overflow downdraw method Methods 0.000 description 2
- 238000012098 association analyses Methods 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 230000005251 gamma ray Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000010355 oscillation Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 238000009827 uniform distribution Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/10—Pre-processing; Data cleansing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
- G06F18/2135—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on approximation criteria, e.g. principal component analysis
-
- 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
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/253—Fusion techniques of extracted features
-
- 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
-
- 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)
- Data Mining & Analysis (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Radio Relay Systems (AREA)
Abstract
The invention discloses a satellite communication radiation source individual identification method based on phase modulation signal feature fusion, which comprises the steps of firstly extracting fingerprint feature mapping matrixes with different mapping forms according to various algorithms, wherein the fingerprint feature mapping matrixes comprise wavelet decomposition features, high-order spectrum features, hausdorff dimension and fractal box dimension; then, fusing the four features obtained in the step 1 by utilizing a CCA algorithm to obtain a new mapping matrix; performing nonlinear dimension reduction processing on the new mapping matrix by using a KPCA algorithm to realize redundancy elimination; and finally, training and identifying the satellite communication radiation source individuals based on the processed new mapping matrix by using a classifier, so as to realize identification and classification of the satellite communication radiation source individuals. The method can fuse the fingerprint features of multiple radiation sources, and based on the nonlinear dimension reduction technology, the fusion features are subjected to redundancy elimination, complex differences inside the satellite transponder units can be mapped more comprehensively, and the method has high robustness.
Description
Technical Field
The invention relates to the technical field of satellite communication and signal processing, in particular to a satellite communication radiation source individual identification method based on phase modulation signal feature fusion.
Background
The satellite communication radiation source individual identification technology is widely applied to the military and civil fields, in a modern military system, satellite communication can ensure the expansion of a series of military operations such as accurate striking, battlefield monitoring, information reconnaissance and the like, has an extremely important role in information battle, and in recent years, the world and the country are developing own satellite communication networks; in civilian terms, positioning, navigation and information transfer of mobile terminals mostly rely on satellite communication networks. The method can realize battlefield real-time situation awareness by carrying out high-precision identification on the satellite communication radiation source individuals, master initiative in electronic countermeasure actions, and has great significance in aspects of civil safety authentication assistance and the like.
The radiation source identification method aiming at the satellite communication scene has been studied, but the scheme in the prior art is to identify the communication equipment in the specific scene based on the single feature, and the mapping capability of the single feature may be insufficient to influence the identification rate when the satellite communication scene meeting various identification requirements is applied in the actual battlefield.
Disclosure of Invention
The invention aims to provide a satellite communication radiation source individual identification method based on phase modulation signal feature fusion, which can fuse various radiation source fingerprint features, and can remove redundancy on the fusion features based on a nonlinear dimension reduction technology, so that complex differences inside a satellite transponder individual can be mapped more comprehensively, and stronger robustness is achieved.
The invention aims at realizing the following technical scheme:
a satellite communication radiation source individual identification method based on phase modulation signal feature fusion, the method comprising:
step 1, firstly extracting fingerprint feature mapping matrixes with different mapping forms according to a plurality of algorithms, wherein the fingerprint feature mapping matrixes comprise wavelet decomposition features, high-order spectrum features, hausdorff dimension and fractal box dimension;
wherein the wavelet decomposition feature is used for mapping oscillator distortion and I/Q mismatch characteristics inside a satellite communication radiation source;
the high-order spectral features are used for mapping nonlinear and non-stable power amplifier distortion and I/Q DC bias characteristics inside a satellite communication radiation source;
the Hausdorff dimension and the fractal box dimension are used for mapping complexity characteristics reflected by a radiation source signal in the satellite communication radiation source;
step 2, fusing the four features obtained in the step 1 by using a typical correlation analysis CCA algorithm to obtain a new mapping matrix;
step 3, performing nonlinear dimension reduction processing on the new mapping matrix by using a kernel principal component analysis KPCA algorithm to realize redundancy elimination;
and step 4, finally training and identifying the satellite communication radiation source individuals based on the new mapping matrix processed in the step 3 by using a classifier, so as to realize identification and classification of the satellite communication radiation source individuals.
According to the technical scheme provided by the invention, the method can fuse the fingerprint features of various radiation sources, and the fused features are subjected to redundancy elimination based on the nonlinear dimension reduction technology, so that the complex differences inside the satellite transponder individuals can be mapped more comprehensively, the robustness is high, the method can be used as a judging basis for identifying the radiation sources in multiple scenes, and the accurate judgment of the satellite communication radiation sources in different scenes is realized.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are needed in the description of the embodiments will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a schematic flow chart of a satellite communication radiation source individual identification method based on phase modulation signal feature fusion provided by an embodiment of the invention;
fig. 2 is a schematic diagram of a fingerprint feature generating module of a satellite communication radiation source according to an embodiment of the invention.
Detailed Description
The technical solutions of 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 apparent that the described embodiments are only some embodiments of the present invention, not all embodiments of the present invention, and this is not limiting to the present invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to fall within the scope of the invention.
Fig. 1 is a schematic flow chart of a satellite communication radiation source individual identification method based on phase modulation signal feature fusion, which is provided by the embodiment of the invention, and the method comprises the following steps:
step 1, firstly extracting fingerprint feature mapping matrixes with different mapping forms according to a plurality of algorithms, wherein the fingerprint feature mapping matrixes comprise wavelet decomposition features, high-order spectrum features, hausdorff dimension and fractal box dimension;
in this step, in the satellite communication radiation source, the fingerprint features are formed from distortions of various modules, such as quadrature modulation distortion caused by direct current bias and IQ mismatch, oscillator distortion caused by frequency offset caused by oscillator error, power amplification distortion caused by working in a nonlinear amplification section, etc., as shown in fig. 2, a schematic diagram of a fingerprint feature generating module of the satellite communication radiation source according to an embodiment of the present invention is shown, and distortion features of each module in the satellite communication radiation source include:
1) Orthogonal modulation distortion: the device comprises an I/Q direct current bias part and an I/Q mismatch part, wherein the I/Q direct current bias is mainly caused by direct current bias of components on an I/Q path and carrier leakage of a mixer, and the I/Q mismatch comprises I/Q gain mismatch and I/Q phase mismatch;
2) Oscillator distortion: the oscillator error causes the signal carrier frequency to shift, thereby introducing frequency offset into the equivalent baseband signal;
3) Power amplifier distortion: ideally, the output amplitude of the power amplifier is linear with the input signal amplitude, but in practice the power amplifier inevitably contains non-linear characteristics.
The I/Q dc offset is represented on the equivalent baseband signal s (t) as:
p(t)＝s(t)+ξ
wherein ζ is complex number, representing DC offset coefficient;
the I/Q gain mismatch is the power inequality caused by the difference of gain characteristics of the amplifiers of the two I/Q paths; the I/Q phase mismatch is that the phase difference caused by the intermediate frequency local oscillation error is not equal to 90 degrees; the I/Q mismatch can be expressed by the following formula:
q(t)＝μp(t)+νp * (t)
wherein μ and v represent parameters of I/Q mismatch generation, respectively;
oscillator distortion can be expressed as:
wherein δ represents the carrier frequency offset; phi (phi) 0 Representing the phase difference between the receiver and the transmitter carrier wave, obeying the uniform distribution from-pi to pi, and needing to be explained is phi 0 Not the fingerprint characteristic of the radiation source, but a random parameter which would exist under ideal conditions.
The power amplifier distortion can be expressed based on the Taylor series as:
wherein M represents the nonlinear order of the power amplifier; gamma ray 2m-1 Is a nonlinear parameter.
In summary, { ζ, μ, ν, δ, γ } and the complexity of the components within the radiation source constitute the main fingerprint characteristics within the satellite communication radiation source.
In this embodiment, for better analysis of the satellite communication radiation source identification system, the following model assumptions are simplified:
1) The channel environment is good enough, and the acquired data set has no excessive clutter;
2) The signal-to-noise ratio of the modulation signal is high enough and is less affected by noise;
3) The number of the data set samples is sufficient, and the classifier can be supported for training;
4) The recognition scene is not open set recognition (i.e., the signal class to be recognized is contained in the training set).
In satellite communication radiation source identification, as the device structure is complex and the signal energy distribution is stray, the vanishing moment of the basis function selected by wavelet decomposition needs to be large, the compression performance of the wavelet basis function can be ensured, and therefore the db1 wavelet basis function is selected for coefficient decomposition; the method adopts the integral double spectrum as the mapping mode of the signal high-order spectrum to map the nonlinear non-stationary characteristic inside the radiation source, because the high-order spectrum redundancy information above the third order is excessive and the calculated amount is extremely large, and the algorithm complexity is influenced; hausdorff dimension and fractal box dimension are two fractal features commonly used in communication radiation source identification and are selected to quantitatively map complexity characteristics inside the radiation source.
In particular, the wavelet decomposition feature is used to map oscillator distortion, I/Q mismatch characteristics inside a satellite communications radiation source; the high-order spectral features are used for mapping nonlinear and non-stable power amplifier distortion and I/Q DC bias characteristics inside a satellite communication radiation source; the Hausdorff dimension and the fractal box dimension are used for mapping complexity characteristics reflected by radiation source signals inside the satellite communication radiation source.
In a specific implementation, the process of extracting the fingerprint feature mapping matrix in different mapping forms according to various algorithms comprises the following steps:
firstThe communication data of the satellite transponder is received, collected and demodulated to obtain an original phase modulation satellite communication signal data set, and the original data set is preprocessed by a preprocessing method based on data enhancement to obtain n modulation signal samples X 0 ＝{x 1 ,x 2 ,…,x n }；
Modulated signal sample X using db1 wavelet basis function ψ (t 0 Wavelet decomposition is performed to obtain wavelet decomposition characteristics Wf expressed as:
wherein a and b respectively represent a scale parameter and a translation parameter; f (t is a single signal sample; t is the coordinates of the signal time axis (hereinafter);
recalculating a modulated signal sample X 0 Third order cumulative amount c of (2) 3f Expressed as:
wherein f * (t) is the conjugate signal of the single signal sample f (t); τ 1 、τ 2 Coordinates on a third-order cumulative volume domain;
for the third order cumulative quantity c 3f Performing two-dimensional discrete time Fourier transform to obtain a two-dimensional matrix characteristic B of a modulated signal sample bispectrum f Expressed as:
wherein omega 1 、ω 2 Is the coordinates on the bispectrum domain;
selecting an integral path and an integral path interval, and performing two-dimensional matrix characteristic B f Performing integral dimension reduction processing to obtain rectangular integral double spectrums of the modulated signal samples, wherein the rectangular integral double spectrums are expressed as follows:
SIBω)＝∮B f (ω 1 ,ω 2 )dω 1 dω 2
then calculating according to fractal theory to obtain a modulation signal sample X 0 Hausdorff dimension D of (v) H Expressed as:
D H ＝inf{p:H p (f)＝0}＝sup{p:H p (f)＝∞}
f is a single signal sample f (t); inf represents the infinit; sup is the upper bound; p is the measure dimension of the sample; h p (f) A Haudorff measure of the sample;
and calculate the fractal box dimension D Box Expressed as:
wherein N is ε (f) Is the minimum number of lattices in the two-dimensional lattice with a maximum diameter epsilon that can cover a single signal sample f (t).
Step 2, fusing the four features obtained in the step 1 by using a typical association analysis (Canonical Correlation Analysis, CCA) algorithm to obtain a new mapping matrix;
in the step, a novel feature mapping mode is constructed for any two feature matrixes X and Y according to the four features obtained in the step 1Wherein ( T Representing a transpose of the matrix; />Representing a mapping matrix; the specific process is as follows:
first, an overall covariance matrix is calculated, expressed as:
the covariance Cov is calculated in a mode of Cov (X, Y) =E (XY) -E (X) and E (Y), wherein E (X) represents mathematical expectation of the matrix;
the lagrangian multiplier is utilized to solve the problem of maximum correlation between two feature matrices X and Y, namely:
where corr represents the correlation between matrices; var represents the matrix variance;
obtaining a linear mapping mode from an original feature matrix to a fusion feature matrix after solving:
and then, two feature matrixes X and Y are connected in series according to a linear mapping mode to obtain a new fusion feature matrix Z, which is expressed as:
then judging whether all four features are fused, if so, finishing feature fusion to obtain a new mapping matrix; otherwise, continuing to fuse the fused feature matrix Z and the unfused feature matrix.
Step 3, performing nonlinear dimension reduction processing on the new mapping matrix by using a kernel principal component analysis (Kennel Principle Component Analysis, KPCA) algorithm to realize redundancy elimination;
in this step, a new mapping matrix is mapped into a high-dimensional space by using a kernel function phi, i.e., a high-dimensional mapping mode;
then a kernel matrix K is defined, expressed as:
K＝φ(X)φ(X) T ＝[k(x i ，x j )] m×m ＝[φ(x i ) T φ(x j )] m×m
wherein phi (X), phi (X) T Is a high-dimensional mapped sample and transpose thereof; m represents the dimension of the kernel matrix K determined by the kernel function phi; x is x i And x j For the purpose of sampleThe i and j elements of the present table.
Defining a centralised kernel matrixExpressed as:
wherein l m Is an m x m matrix with matrix elements of 1/m;
based on a centralised matrix of nucleiPerforming eigenvalue decomposition on a new mapping matrix in a high-dimensional space, selecting dimension r of dimension reduction, and arranging from large to small according to eigenvalues; selecting the first r corresponding feature vectors to obtain a mapping matrix after dimension reduction;
the reduced-dimension mapping matrix can be used for classifier training.
And step 4, finally training and identifying the satellite communication radiation source individuals based on the new mapping matrix processed in the step 3 by using a classifier, so as to realize identification and classification of the satellite communication radiation source individuals.
In the step, the new mapping matrix and the corresponding class label processed in the step 3 are input into a classifier for training, and the obtained model is used for individual identification of a satellite communication radiation source, specifically:
for the new mapping matrix Z, it is divided into test sets C 1 And training set C 2 And give corresponding class label L 1 And L is equal to 2 For training set C 2 Model training is carried out:
H＝ζ(C 2 ，L 2 )
wherein H is based on training set C 2 Category label L 2 A determined classifier model; zeta is a K-Nearest Neighbor (KNN) recognition algorithm;
in the identification process, calculating an unknown sample point x to be identified a In the sample space (i.e. C 2 ) From other onesSample point x b Is expressed as:
dist＝|x a -x b |
wherein, |represents modulo arithmetic; arranging samples from small to large according to Euclidean distance dist, and taking the first q (q is artificially selected neighbor number) sample points { x } 1 ，x 2 ，...，x q Using the same as q nearest neighbor samples, counting the category with the highest occurrence number as a sample point x a Is a category of (2);
after the identification is finished, the sample point x is obtained a Respectively adding the recognition results into the training set C 2 And category label L 2 Forming a new training set and a new class label;
test set C for classifier model H 1 Testing is carried out to obtain a test result L', which is expressed as:
L′＝H(C 1 )
test result L' and class label L 1 A comparison is made to verify the accuracy.
It is noted that what is not described in detail in the embodiments of the present invention belongs to the prior art known to those skilled in the art.
In summary, the method according to the embodiment of the present invention can accurately obtain the satellite communication modulation signal received from the site without prior information about the equipment model and the channel environment; and the CCA fusion method is combined with the KPCA dimension reduction method, the CCA fusion method integrates various fingerprint feature mapping means, the individual difference inside the satellite radiation source can be reflected more comprehensively, the robustness of the identification system is effectively improved, the effectiveness of the identification algorithm under a multi-identification scene is ensured, the KPCA dimension reduction method ensures that the complexity of the algorithm is not excessive, and therefore the individual accurate identification of the satellite communication radiation source is realized.
In addition, it will be understood by those skilled in the art that all or part of the steps in implementing the methods of the above embodiments may be implemented by a program to instruct related hardware, and the corresponding program may be stored in a computer readable storage medium, where the storage medium may be a read only memory, a magnetic disk or an optical disk, etc.
The foregoing is only a preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any changes or substitutions easily contemplated by those skilled in the art within the scope of the present invention should be included in the scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims. The information disclosed in the background section herein is only for enhancement of understanding of the general background of the invention and is not to be taken as an admission or any form of suggestion that this information forms the prior art already known to those of ordinary skill in the art.
Claims (4)
1. A satellite communication radiation source individual identification method based on phase modulation signal feature fusion, which is characterized by comprising the following steps:
step 1, firstly extracting fingerprint feature mapping matrixes with different mapping forms according to a plurality of algorithms, wherein the fingerprint feature mapping matrixes comprise wavelet decomposition features, high-order spectrum features, hausdorff dimension and fractal box dimension;
wherein the wavelet decomposition feature is used for mapping oscillator distortion and I/Q mismatch characteristics inside a satellite communication radiation source;
the high-order spectral features are used for mapping nonlinear and non-stable power amplifier distortion and I/Q DC bias characteristics inside a satellite communication radiation source;
the Hausdorff dimension and the fractal box dimension are used for mapping complexity characteristics reflected by a radiation source signal in the satellite communication radiation source;
step 2, fusing the four features obtained in the step 1 by using a typical correlation analysis CCA algorithm to obtain a new mapping matrix;
the process of the step 2 specifically comprises the following steps:
aiming at the four features obtained in the step 1, a novel feature mapping mode is constructed for any two feature matrixes X and YWherein () T Representing a transpose of the matrix; />Representing a mapping matrix; the specific process is as follows:
first, an overall covariance matrix is calculated, expressed as:
the covariance Cov is calculated in a mode of Cov (X, Y) =E (XY) -E (X) and E (Y), wherein E (X) represents mathematical expectation of the matrix;
the lagrangian multiplier is utilized to solve the problem of maximum correlation between two feature matrices X and Y, namely:
where corr represents the correlation between matrices; var represents the matrix variance;
obtaining a linear mapping mode from an original feature matrix to a fusion feature matrix after solving:
and then, two feature matrixes X and Y are connected in series according to a linear mapping mode to obtain a new fusion feature matrix Z, which is expressed as:
then judging whether all four features are fused, if so, finishing feature fusion to obtain a new mapping matrix; otherwise, continuing to fuse the fused feature matrix Z and the unfused feature matrix;
step 3, performing nonlinear dimension reduction processing on the new mapping matrix by using a kernel principal component analysis KPCA algorithm to realize redundancy elimination;
and step 4, finally training and identifying the satellite communication radiation source individuals based on the new mapping matrix processed in the step 3 by using a classifier, so as to realize identification and classification of the satellite communication radiation source individuals.
2. The method for identifying an individual satellite communication radiation source based on phase modulation signal feature fusion according to claim 1, wherein in step 1, the process of extracting fingerprint feature mapping matrices in different mapping forms according to a plurality of algorithms is specifically as follows:
firstly, receiving, collecting and demodulating communication data of a satellite transponder to obtain an original phase modulation satellite communication signal data set, and preprocessing the original data set by utilizing a preprocessing method based on data enhancement to obtain n modulation signal samples X 0 ＝{x 1 ,x 2 ,…,x n }；
Modulated signal sample X using db1 wavelet basis function ψ (t) 0 Wavelet decomposition is performed to obtain wavelet decomposition characteristics Wf expressed as:
wherein a and b respectively represent a scale parameter and a translation parameter; f (t) is a single signal sample; t is the coordinates of a signal time axis;
recalculating a modulated signal sample X 0 Third order cumulative amount c of (2) 3f Expressed as:
wherein f * (t) is the conjugate signal of the single signal sample f (t); τ 1 、τ 2 Coordinates on a third-order cumulative volume domain;
for the third order cumulative quantity c 3f Performing two-dimensional discrete time Fourier transform to obtain a modulated signal sample bispectrumTwo-dimensional matrix feature B f Expressed as:
wherein omega 1 、ω 2 Is the coordinates on the bispectrum domain;
selecting an integral path and an integral path interval, and performing two-dimensional matrix characteristic B f Performing integral dimension reduction processing to obtain rectangular integral double spectrums of the modulated signal samples, wherein the rectangular integral double spectrums are expressed as follows:
SIB(ω)＝∮B f (ω 1 ,ω 2 )dω 1 dω 2
then calculating according to fractal theory to obtain a modulation signal sample X 0 Hausdorff dimension D of (v) H Expressed as:
D H ＝inf{p：H p (f)＝0}＝sup{p：H p (f)＝∞}
f is a single signal sample f (t); inf represents the infinit; sup is the upper bound; p is the measure dimension of the sample; h p (f) A Haudorff measure of the sample;
and calculate the fractal box dimension D Box Expressed as:
wherein N is ε (f) Is the minimum number of lattices in the two-dimensional lattice with a maximum diameter epsilon that can cover a single signal sample f (t).
3. The method for identifying the satellite communication radiation source based on the phase modulation signal feature fusion according to claim 1, wherein the process of the step 3 specifically comprises the following steps:
firstly, mapping a new mapping matrix into a high-dimensional space by using a kernel function phi, namely a high-dimensional mapping mode;
then a kernel matrix K is defined, expressed as:
K＝φ(X)φ(X) T ＝[k(x i ，x j )] m×m ＝[φ(x i ) T φ(x j )] m×m
wherein phi (X), phi (X) T Is a high-dimensional mapped sample and transpose thereof; m represents the dimension of the kernel matrix K determined by the kernel function phi; x is x i And x j I, j elements of the sample;
defining a centralised kernel matrixExpressed as:
wherein l m Is an m x m matrix with matrix elements of 1/m;
based on a centralised matrix of nucleiPerforming eigenvalue decomposition on a new mapping matrix in a high-dimensional space, selecting dimension r of dimension reduction, and arranging from large to small according to eigenvalues; selecting the first r corresponding feature vectors to obtain a mapping matrix after dimension reduction;
the reduced-dimension mapping matrix can be used for classifier training.
4. The method for identifying the individual satellite communication radiation source based on the phase modulation signal feature fusion according to claim 1, wherein in step 4, specifically, the new mapping matrix and the corresponding class label processed in step 3 are input into a classifier for training, and the obtained model is used for identifying the individual satellite communication radiation source, specifically:
for the new mapping matrix Z, it is divided into test sets C 1 And training set C 2 And give corresponding class label L 1 And L is equal to 2 For training set C 2 Model training is carried out:
H＝ζ(C 2 ，L 2 )
wherein H is based on training set C 2 Category label L 2 A determined classifier model; ζ is a K neighbor recognition algorithm selected;
in the identification process, calculating an unknown sample point x to be identified a X from other sample points in sample space b Is expressed as:
dist＝|x a -x b |
wherein, |represents modulo arithmetic; the samples are arranged from small to large according to Euclidean distance dist, and the first q sample points { x } 1 ,x 2 ,…,x q Used as q nearest neighbor samples, and the category with the highest occurrence number is counted as a sample point x a Is a category of (2);
after the identification is finished, the sample point x is obtained a Respectively adding the recognition results into the training set C 2 And category label L 2 Forming a new training set and a new class label;
test set C for classifier model H 1 Testing is carried out to obtain a test result L', which is expressed as:
L′＝H(C 1 )
test result L' and class label L 1 A comparison is made to verify the accuracy.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310479126.0A CN116431991B (en) | 2023-04-27 | 2023-04-27 | Satellite communication radiation source individual identification method based on phase modulation signal feature fusion |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310479126.0A CN116431991B (en) | 2023-04-27 | 2023-04-27 | Satellite communication radiation source individual identification method based on phase modulation signal feature fusion |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116431991A CN116431991A (en) | 2023-07-14 |
| CN116431991B true CN116431991B (en) | 2023-11-28 |
Family
ID=87092662
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310479126.0A Active CN116431991B (en) | 2023-04-27 | 2023-04-27 | Satellite communication radiation source individual identification method based on phase modulation signal feature fusion |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116431991B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105933006A (en) * | 2016-06-24 | 2016-09-07 | 中国科学技术大学 | Single-bit compression sampling method based on time-varying threshold |
| CN110197209A (en) * | 2019-05-15 | 2019-09-03 | 电子科技大学 | A kind of Emitter Recognition based on multi-feature fusion |
-
2023
- 2023-04-27 CN CN202310479126.0A patent/CN116431991B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105933006A (en) * | 2016-06-24 | 2016-09-07 | 中国科学技术大学 | Single-bit compression sampling method based on time-varying threshold |
| CN110197209A (en) * | 2019-05-15 | 2019-09-03 | 电子科技大学 | A kind of Emitter Recognition based on multi-feature fusion |
Non-Patent Citations (7)
| Title |
|---|
| Redshift space distortions in Lagrangian space and the Linear large scale velocity field of dark matter;Emily Tyhurst等;《https://arxiv.org/abs/2202.08435》;1-9 * |
| 基于信号分解的无线网络设备个体特征提取方法;马建明;《中国优秀硕士学位论文全文数据库 信息科技辑》;第1.2.1节、第3.4.2节、第4.2节、图1-1 * |
| 基于信号指纹的通信设备个体识别研究;范友源;《中国优秀硕士学位论文全文数据库 信息科技辑》;第2.2.2节、第3.4节、第4.2.2节 * |
| 基于光纤信道的量子网络的若干问题研究;江扬帆;《中国博士学位论文全文数据库 基础科学辑》;A005-14 * |
| 平面电磁波在非均匀等离子体柱上的散射研究;吴小坡等;《真空科学与技术学报》;1003-1008 * |
| 李秋生. 分数阶Fourier域常规雷达目标识别的分形理论及其应用.西安电子科技大学出版社,2022,49-51. * |
| 马建明.基于信号分解的无线网络设备个体特征提取方法.《中国优秀硕士学位论文全文数据库 信息科技辑》.2021,第1.2.1节、第3.4.2节、第4.2节、图1-1. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116431991A (en) | 2023-07-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Gok et al. | A new method for specific emitter identification with results on real radar measurements | |
| CN109307862A (en) | A kind of target radiation source individual discrimination method | |
| CN108375752B (en) | Amplitude-phase error single-radiation-source direction finding method based on all-angle search | |
| CN111050315B (en) | Wireless transmitter identification method based on multi-core two-way network | |
| CN101959191B (en) | Safety authentication method and system for wireless network | |
| CN110197209A (en) | A kind of Emitter Recognition based on multi-feature fusion | |
| Wu et al. | Specific emitter identification for satellite communication using probabilistic neural networks | |
| CN111428817A (en) | Defense method for resisting attack by radio signal identification | |
| CN114896887B (en) | Frequency-using equipment radio frequency fingerprint identification method based on deep learning | |
| CN106842198A (en) | A kind of subband fusion of imaging algorithm based on convex optimum theory | |
| CN106405543A (en) | Cognition-type blind source separation radiant source extraction method and evaluation method thereof | |
| CN115661869A (en) | Radio frequency signal fingerprint identification method and system based on multi-dimensional feature intelligent fusion | |
| CN101252382A (en) | Wide frequency range signal polarizing and DOA estimating method and apparatus | |
| CN116431991B (en) | Satellite communication radiation source individual identification method based on phase modulation signal feature fusion | |
| CN115186714B (en) | Network card frequency spectrum fingerprint feature amplification method based on feature correlation and self-adaptive decomposition | |
| Cun et al. | Specific emitter identification based on eye diagram | |
| CN111222430A (en) | Unmanned aerial vehicle identification method and system based on artificial intelligence | |
| CN116628459A (en) | Unmanned aerial vehicle remote control signal identification method and system based on four-dimensional code | |
| Patton et al. | Autocorrelation and modulus constraints in radar waveform optimization | |
| Zhang et al. | Specific emitter identification based on power amplifier | |
| CN102054273B (en) | Simplex triangular decomposition-based method for decomposing mixed pixels of hyperspectral remote sensing images | |
| Feng et al. | FCGCN: Feature Correlation Graph Convolution Network for Few-Shot Individual Identification | |
| Sun et al. | Radio Frequency Fingerprinting based on Differential Reconstructed Phase Space | |
| Su et al. | Parameter estimation of linear frequency modulated signals based on a Wigner–Ville distribution complex-valued convolutional neural network | |
| Ran et al. | Mobilenet for differential constellation trace figure |
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 |