CN112213698B - Deception jamming identification method based on sparse representation classification - Google Patents
Deception jamming identification method based on sparse representation classification Download PDFInfo
- Publication number
- CN112213698B CN112213698B CN202011072188.2A CN202011072188A CN112213698B CN 112213698 B CN112213698 B CN 112213698B CN 202011072188 A CN202011072188 A CN 202011072188A CN 112213698 B CN112213698 B CN 112213698B
- Authority
- CN
- China
- Prior art keywords
- signal
- signals
- interference
- frequency bands
- deception jamming
- 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
- 238000000034 method Methods 0.000 title claims abstract description 34
- 238000000354 decomposition reaction Methods 0.000 claims abstract description 7
- 238000007500 overflow downdraw method Methods 0.000 claims abstract description 4
- 239000011159 matrix material Substances 0.000 claims description 18
- 239000013598 vector Substances 0.000 claims description 18
- 238000003672 processing method Methods 0.000 claims description 4
- 230000004927 fusion Effects 0.000 claims description 3
- 238000009825 accumulation Methods 0.000 claims 3
- 238000004364 calculation method Methods 0.000 abstract description 2
- 230000001186 cumulative effect Effects 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000001427 coherent effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 230000015654 memory Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000035485 pulse pressure Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
- G01S7/02—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00
- G01S7/36—Means for anti-jamming, e.g. ECCM, i.e. electronic counter-counter measures
-
- 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)
- Computer Networks & Wireless Communication (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Radar Systems Or Details Thereof (AREA)
Abstract
Aiming at the current situation that the identification of radar deception jamming signals is difficult, the invention provides a deception jamming identification method based on sparse representation classification, which utilizes the variation difference of deception jamming signal parameters to realize the identification of several common deception jamming signals, and receives signals of different slow time domains to form a signal set by dividing frequency bands; performing third-order cumulant calculation on signals of each slow time domain moment in different frequency bands, extracting cumulant slice characteristics, further reducing the influence of noise, and extracting stable signal characteristics; and (3) reducing the dimension of the features by utilizing singular value decomposition, extracting main components, classifying and identifying signals on different frequency bands by utilizing sparse representation classification, and integrating classification results by utilizing a decision fusion method. The invention can fully utilize the change information of the parameters of different deception jamming signals in the towing period, and effectively identify several common deception jamming signals.
Description
Technical Field
The invention relates to the technical field of radar deception jamming recognition, in particular to a deception jamming recognition method based on sparse representation classification.
Background
With the application of digital radio frequency storage (Digital Radio Frequency Memories, DRFM) technology, a DRFM jammer can rapidly capture radar emission signals, a sampling device with high frequency is arranged in the DRFM jammer, the radar emission signals can be recovered with high fidelity and modulated to generate interference signals with deception properties, and the deception interference signals and radar echo signals are highly coherent, so that the same pulse pressure gain is obtained with the radar echo signals when the radar receiver is entered, the radar cannot distinguish true signals from false signals, and the radar is a main threat of the radar.
Most radar deception jamming recognition algorithms at present do not utilize the change information of signals in slow time domain parameters, are greatly affected by noise, and are difficult to maintain high recognition rate under low signal-to-noise ratio; the partial fraud recognition method can only detect whether the radar is interfered by fraud, and cannot detect which specific type the interfered belongs to.
Disclosure of Invention
According to the above, the main technical purpose of the present invention is to provide a sparse representation classification-based spoofing interference recognition method, which uses the variation difference of spoofing interference signal parameters to extract the parameter variation rules of signals in different slow time domains, so as to recognize the specific types of the signals and extract the characteristics of different kinds of signals.
A deception jamming recognition method based on sparse representation classification comprises the following steps:
step 1, forming a signal set by receiving a plurality of signals in a slow time domain, carrying out frequency division on the formed signal set, reconstructing the signals, identifying deception interference signals by utilizing different change modes of signal parameters on different frequency bands, and classifying and identifying received signals on different frequency bands by utilizing information differences among the signals on different frequency bands;
step 2, signal processing is carried out on the signal sets of different frequency bands on each slow time domain, stable signal characteristics are extracted, the influence of noise on an identification system is reduced, the difference between different types of signals is highlighted, and a characteristic matrix for identification is constructed on each frequency band;
step 3, performing dimension reduction processing on the feature matrix on each frequency band, and replacing the information of the whole feature matrix with a small amount of feature vectors so as to reduce the number of feature parameters in the subsequent identification process;
and 4, identifying different deception jamming signals by using a sparse representation classification method, taking a large amount of sample data, taking feature vectors after dimension reduction as atoms, constructing an overcomplete dictionary, expanding the samples on the dictionary, performing deception jamming identification on different divided frequency bands, integrating classification results on different frequency bands by using a decision fusion method, and obtaining a final identification result.
Preferably, in the step 1, the signal at the slow time domain moment is a chirp signal, and the transmission signal form of the signal is:
wherein:wherein f 0 For intermediate frequency, k is modulation slope, +.>For the initial phase of the transmitted signal, τ is the pulse width.
Preferably, in the step 1, the spoofing interference signal includes a signal form including a distance towing interference, a speed towing interference, and a distance-speed combined towing interference, where the distance towing interference form is as follows:
wherein: a is that R To the amplitude of the range-drag disturbance, Δt J Inherent delay, Δt, required for jammers to receive, store, process, forward radar signals J (t) is the modulation delay from the trailing interference,the method comprises the steps of (1) setting an initial phase of an interference signal of an jammer;
the speed towing interference is as follows:
wherein: a is that V For the amplitude of the speed-trailing disturbance Δf dJ (t) doppler shift for velocity trailing interference modulation;
the range-speed joint trailing interference is in the form of:
wherein: a is that R-V The magnitude of the disturbance is dragged for the range-speed combination.
Preferably, in the step 1, the method for dividing different frequency bands adopts wavelet packet reconstruction, and after frequency band division, the obtained signal set adopts the following form:
S i ＝[x i (t,η 0 ),x i (t,η 1 ),…,x i (t,η n ),…x i (t,η N )] T
the radar receiver is slow in the time domain eta at the ith frequency band n Time-of-day received signal x i (t,η n ) There are several cases:
wherein F represents a Fourier transform, R 0 (η n ) Is slow time domain eta n Instant distance, Δt, corresponding to moment J (t,η n ) Is slow time domain eta n Delay amount of time-of-day distance spoofing interference, Δf dJ (t,η n ) Is slow time domain eta n Doppler frequency shift H of speed dragging interference signals generated by digital radio frequency storage jammers corresponding to time i (ω) represents the corresponding band pass filter in the ith band.
Preferably, in the step 2, the processing method of the signal feature uses a third-order cumulative amount, and the cumulative amount is extracted from the diagonal slices, where the third-order cumulative amount uses the following form:
wherein τ 1 、τ 2 Delay amounts respectively, and E represents mathematical expectation;
diagonal slices take the following form:
preferably, in the step 3, the dimension reduction processing method of the feature matrix adopts a singular value decomposition method, and the left and right singular vectors corresponding to the singular values are extracted as feature vectors for subsequent recognition.
Preferably, in the step 4, the atoms used for sparse representation classification are singular vectors, and the decision fusion takes the following form:
y t ＝argmax([y 1 ,y 2 ,y 3 ,y 4 ])
wherein y is 1 、y 2 、y 3 、y 4 The dictionary numbers of the target echo and the deception jamming signals are respectively represented, and the final classification result is represented as the signal type with the largest classification times after the dictionary is identified.
Compared with the prior art, the invention has the beneficial effects that:
(1) The novel deception jamming recognition method can recognize common deception jamming signals, is stable in recognition effect and still has high recognition rate under low signal-to-noise ratio.
(2) The signal parameter information of different slow time domains is fully utilized, and the deception jamming signal is essentially identified by utilizing the variation difference of deception jamming signal parameters.
Drawings
FIG. 1 is a partial step explanatory diagram of a sparse representation classification-based fraud identification method;
fig. 2 is an algorithm flow chart of a sparse representation classification based fraud recognition method.
Detailed Description
The invention is further illustrated below in connection with specific embodiments.
Examples
Referring to fig. 1-2, the method for identifying spoofing interference based on sparse representation classification provided by the invention comprises the following steps:
(1) Frequency band division is carried out on the received signals, and a signal set of different frequency bands is formed by combining a slow time domain, wherein the modeling of the signals is as follows:
the radar transmitting signal is a linear frequency modulation signal, and the transmitting signal is in the form of:
wherein:f 0 for intermediate frequency, k is modulation slope, +.>For the initial phase of the transmitted signal, τ is the pulse width.
The deception jamming signal has a range towing jamming signal form, a speed towing deception jamming signal form and a range-speed combined towing jamming signal form, wherein the range towing jamming form is as follows:
wherein: a is that R To the amplitude of the range-drag disturbance, Δt J Inherent delay, Δt, required for jammers to receive, store, process, forward radar signals J (t) is the modulation delay from the trailing interference,the initial phase of the jammer signal.
The speed towing interference is as follows:
wherein: a is that V For the amplitude of the speed-trailing disturbance Δf dJ And (t) is the Doppler shift of the velocity trailing interference modulation.
The range-speed joint trailing interference is in the form of:
wherein: a is that R-V The magnitude of the disturbance is dragged for the range-speed combination.
Carrying out wavelet packet decomposition and reconstruction on the received signals to obtain signal sets on different frequency bands, wherein the signal sets are in the following forms:
S i ＝[x i (t,η 0 ),x i (t,η 1 ),…,x i (t,η n ),…x i (t,η N )] T
wherein S is i Representing the signal set on the ith wavelet decomposition band, i=1, 2, …,8, η n Represents the slow time domain, n=1, 2, …, N, x i (t,η n ) For the slow time domain eta on the ith frequency band n The signal vector received at the moment corresponds to different deception jamming signals and receives a signal x i (t,η n ) There are several cases:
(2) Extracting the characteristics of signals in different frequency bands to obtain characteristic matrixes in different frequency bands, wherein the specific process is as follows:
and carrying out third-order cumulant operation on signals at different slow time domain moments on each frequency band, wherein the third-order cumulant operation is defined as follows:
wherein τ 1 、τ 2 The delay amounts, respectively, E represent mathematical expectations.
After the third-order cumulant operation, the diagonal slices of the cumulant are extracted, and the obtained characteristics can be expressed as follows:
for reception on the ith frequency bandSet of incoming signals S i And (3) performing third-order cumulant calculation on each slow time domain moment to obtain stable variation characteristics of signals on an ith frequency band, wherein the obtained characteristic matrix can be written into the following form:
A i ＝[f i (τ′,η 0 ),f i (τ′,η 1 ),…,f i (τ′,η n ),…f i (τ′,η N )] T
(3) The feature matrix is subjected to dimension reduction, feature redundancy is reduced, and feature vectors of each frequency band are extracted, wherein the specific process is as follows:
for each frequency band, a feature matrix A with the size of m multiplied by n i Where m represents the length of the slow time domain and n is the received cumulative slice length. According to singular value decomposition, there are
A i ＝U i Σ i V i T
Wherein U is i ＝[u i1 ,u i2 ,u i3 ,…,u ik ,…,u im ]Is A i Left singular matrix of u ik K=1, 2,3, …, m for the m×1 left singular vector. V (V) i ＝[v i1 ,v i2 ,v i3 ,…,v il ,…,v in ]As matrix A i Right singular matrix of v il Is an n×1 right singular vector, l=1, 2,3, …, n.
σ ij Representation matrix A i Is a singular value of (a), reflects the inherent characteristics of the matrix, where j=1, 2,3, …, min (m, n), and satisfies σ i1 ＞σ i2 ＞σ i3 ＞…＞σ imin(m,n) . Thus A is i Can be written as
Due to the target echo signal belonging to a single unitThe component signals and the target echo and spoof interfering signal structures have similarities, so that the first singular value in the singular value distribution of the feature matrix is much greater in value than the other singular values. The feature matrix A can be obtained by extracting the maximum singular value and the singular vector corresponding to the maximum singular value i Overall information.
(4) Expanding the sample on an overcomplete dictionary on a corresponding frequency band to obtain a classification result on the corresponding frequency band, and integrating the results by a decision fusion method to obtain a final recognition result of the sample, wherein the specific process is as follows:
by collecting a large number of samples, extracting feature vectors on corresponding frequency bands as atoms of a dictionary, an overcomplete dictionary is constructed, and according to the sparse representation classification principle, the linear relationship among similar samples is strong, and the linear relationship among different types of samples is poor, so that when the overcomplete dictionary is used for representing an unknown sample, the result is often easily represented as the linear combination of similar atoms in the dictionary. Set D i ＝[d i,1 ,d i,2 ,…,d i,j ,…,d i,mi ]Is a set of samples of type i, where d i,j ∈R n N represents the dimension of the vector, m i Indicating the number of samples of class i. If set D i Is overcomplete, then either does not belong to set D i I-th class sample y of (2) i Can all use D i The elements of (a) represent:
thus, y i Can pass through set D i Expressed by, i.e. y i ＝D i x i WhereinIs sample y i At set D i The linearity formed represents the coefficient. Existing d= [ D 1 ,D 2 ,…,D i ,…,D M ]Where M represents the total number of categories and D is an overcomplete set of samples of each category, referred to as a dictionary.Given a sample y E R to be measured n And assuming it belongs to class i, the sample can be spread out over D:
y＝Dx 0
in the method, in the process of the invention,is a sparse representation of sample y formed on dictionary D. X is x 0 Corresponding to the i-th class atom in the non-zero term dictionary D to obtain x 0 Non-zero term coefficients of (2) can be obtained by solving for l 1 The minimum value of the regular expression converts the problem into solving the optimization problem:
||·|| 1 representation l 1 Norm, epsilon, represents the upper error limit. In practice, the sample data tends to be subject to noise,the locations of non-zero term coefficients of (a) are not all concentrated on dictionary class i atoms. In order to classify, the error amount between the sample and the reconstructed sample on each class sub-dictionary needs to be calculated, and the smallest error is the final classification result:
in the method, in the process of the invention,representation sample y at sparse representation coefficient +.>And (3) setting the rest coefficients to zero according to the coefficients expressed on the i-th type atoms, and obtaining the error between the sample reconstructed by each sub-dictionary and the real sample. The final classification result is as follows:
for the classification result on each frequency band, integration is needed in a decision fusion mode, and the final result can be expressed as:
y t ＝argmax([y 1 ,y 2 ,y 3 ,y 4 ])
wherein y is 1 、y 2 、y 3 、y 4 And respectively representing the dictionary numbers of target echo, RGPO, VGPO and R-VGPO, wherein the integrated recognition result is the signal type with the largest occurrence number after the recognition of each dictionary.
A flowchart of the overall recognition algorithm is shown in fig. 2. The recognition flow can be expressed as a linear decomposition process, when the parameter change rule of the sample is matched with atoms in the dictionary, the parameter change rule of the two signals is proved to be similar, and the final recognition result is often the signal category which is matched with the pattern of the final recognition result. Therefore, the identification method can finish the classification and identification of various deception jamming signals from the aspect of the change rule of the signal parameters, and has high applicability.
The foregoing is only a preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art, who is within the scope of the present invention, should make equivalent substitutions or modifications according to the technical scheme of the present invention and the inventive concept thereof, and should be covered by the scope of the present invention.
Claims (1)
1. The deception jamming recognition method based on sparse representation classification is characterized by comprising the following steps of:
step 1, forming a signal set by receiving a plurality of signals in a slow time domain, carrying out frequency division on the formed signal set, reconstructing the signals, identifying deception interference signals by utilizing different change modes of signal parameters on different frequency bands, and classifying and identifying received signals on different frequency bands by utilizing information differences among the signals on different frequency bands;
step 2, signal processing is carried out on the signal sets of different frequency bands on each slow time domain, stable signal characteristics are extracted, the influence of noise on an identification system is reduced, the difference between different types of signals is highlighted, and a characteristic matrix for identification is constructed on each frequency band;
step 3, performing dimension reduction processing on the feature matrix on each frequency band, and replacing the information of the whole feature matrix with a small amount of feature vectors so as to reduce the number of feature parameters in the subsequent identification process;
step 4, identifying different deception jamming signals by using a sparse representation classification method, taking a large amount of sample data, taking feature vectors after dimension reduction as atoms, constructing an overcomplete dictionary, expanding the samples on the dictionary, carrying out deception jamming identification on different divided frequency bands, integrating classification results on different frequency bands by using a decision fusion method, and obtaining a final identification result;
in the step 1, the signal at the slow time domain moment is a linear frequency modulation signal, and the transmitting signal form of the signal is:
wherein:wherein f 0 For intermediate frequency, k is modulation slope, +.>For the initial phase of the transmitted signal, τ is the pulse width;
in the step 1, the spoofing interference signal includes a signal form including a distance towing interference, a speed towing interference and a distance-speed combined towing interference, wherein the distance towing interference form is as follows:
wherein: a is that R To the amplitude of the range-drag disturbance, Δt J Inherent delay, Δt, required for jammers to receive, store, process, forward radar signals J (t) is the modulation delay from the trailing interference,the method comprises the steps of (1) setting an initial phase of an interference signal of an jammer;
the speed towing interference is as follows:
wherein: a is that V For the amplitude of the speed-trailing disturbance Δf dJ (t) doppler shift for velocity trailing interference modulation;
the range-speed joint trailing interference is in the form of:
wherein: a is that R-V The magnitude of the disturbance is dragged for the distance-speed combination;
in the step 1, the dividing method of different frequency bands adopts wavelet packet reconstruction, and after the frequency bands are divided, the obtained signal set adopts the following form:
S i ＝[x i (t,η 0 ),x i (t,η 1 ),…,x i (t,η n ),…x i (t,η N )] T
the radar receiver is slow in the time domain eta at the ith frequency band n Time-of-day received signal x i (t,η n ) There are several cases:
wherein F represents a Fourier transform, R 0 (η n ) Is slow time domain eta n Instant distance, Δt, corresponding to moment J (t,η n ) Is slow time domain eta n Delay amount of time-of-day distance spoofing interference, Δf dJ (t,η n ) Is slow time domain eta n Doppler frequency shift H of speed dragging interference signals generated by digital radio frequency storage jammers corresponding to time i (ω) represents the corresponding band pass filter in the ith frequency band;
in the step 2, the signal feature processing method adopts a third-order accumulation amount, and the diagonal slice of the accumulation amount is extracted, wherein the third-order accumulation amount adopts the following form:
wherein τ 1 、τ 2 Delay amounts respectively, and E represents mathematical expectation;
diagonal slices take the following form:
in the step 3, the feature matrix dimension reduction processing method adopts a singular value decomposition method, and a left singular vector and a right singular vector corresponding to singular values are extracted to serve as feature vectors for subsequent identification;
in the step 4, the atoms used for sparse representation classification are singular vectors, and the decision fusion adopts the following form:
y t ＝arg max([y 1 ,y 2 ,y 3 ,y 4 ])
wherein y is 1 、y 2 、y 3 、y 4 The dictionary numbers of the target echo and the deception jamming signals are respectively represented, and the final classification result is represented as the signal type with the largest classification times after the dictionary is identified.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011072188.2A CN112213698B (en) | 2020-10-09 | 2020-10-09 | Deception jamming identification method based on sparse representation classification |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011072188.2A CN112213698B (en) | 2020-10-09 | 2020-10-09 | Deception jamming identification method based on sparse representation classification |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112213698A CN112213698A (en) | 2021-01-12 |
| CN112213698B true CN112213698B (en) | 2024-03-29 |
Family
ID=74052936
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011072188.2A Active CN112213698B (en) | 2020-10-09 | 2020-10-09 | Deception jamming identification method based on sparse representation classification |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112213698B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114722710B (en) * | 2022-04-11 | 2023-02-24 | 电子科技大学 | Range gate dragging interference method based on random simulation optimization |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103837863A (en) * | 2014-03-05 | 2014-06-04 | 中国人民解放军海军航空工程学院 | Distance-speed synchronous pull-off deception jamming recognition algorithm based on gradient projection |
| CN107918110A (en) * | 2017-11-06 | 2018-04-17 | 西安电子科技大学 | Linear FM signal deception jamming recognition methods based on multiple resemblance Coefficient |
| CN110208830A (en) * | 2019-07-16 | 2019-09-06 | 东北大学秦皇岛分校 | A kind of navigation anti-interference method based on space-time two-dimensional thinned array |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9562968B2 (en) * | 2012-10-22 | 2017-02-07 | Saab-Sensis Corporation | Sensor system and method for determining target location using sparsity-based processing |
-
2020
- 2020-10-09 CN CN202011072188.2A patent/CN112213698B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103837863A (en) * | 2014-03-05 | 2014-06-04 | 中国人民解放军海军航空工程学院 | Distance-speed synchronous pull-off deception jamming recognition algorithm based on gradient projection |
| CN107918110A (en) * | 2017-11-06 | 2018-04-17 | 西安电子科技大学 | Linear FM signal deception jamming recognition methods based on multiple resemblance Coefficient |
| CN110208830A (en) * | 2019-07-16 | 2019-09-06 | 东北大学秦皇岛分校 | A kind of navigation anti-interference method based on space-time two-dimensional thinned array |
Non-Patent Citations (1)
| Title |
|---|
| 基于栈式稀疏自编码器的有源欺骗干扰识别;阮怀林;杨兴宇;;探测与控制学报;20180826(第04期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112213698A (en) | 2021-01-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105866749B (en) | The radar chaff recognition methods that a kind of distance and speed sync tow | |
| CN112882009B (en) | Radar micro Doppler target identification method based on amplitude and phase dual-channel network | |
| CN112560803A (en) | Radar signal modulation identification method based on time-frequency analysis and machine learning | |
| CN109738887B (en) | Target human motion state identification method suitable for micro-motion interference scene | |
| CN102510363A (en) | LFM (linear frequency modulation) signal detecting method under strong interference source environment | |
| CN111050315B (en) | Wireless transmitter identification method based on multi-core two-way network | |
| CN116359851A (en) | Radar active interference detection and identification method and device based on converged network | |
| CN113640768B (en) | Low-resolution radar target identification method based on wavelet transformation | |
| CN108169728A (en) | Range extension target detection method based on Minkowski distances | |
| CN116047427B (en) | Small sample radar active interference identification method | |
| CN112213698B (en) | Deception jamming identification method based on sparse representation classification | |
| CN112213697A (en) | Feature fusion method for radar deception jamming recognition based on Bayesian decision theory | |
| Mehr et al. | Detection and classification of GNSS jammers using convolutional neural networks | |
| Chen et al. | Suppressive interference suppression for airborne SAR using BSS for singular value and eigenvalue decomposition based on information entropy | |
| CN109472239B (en) | Individual identification method of frequency hopping radio station | |
| Zhang et al. | Clutter Suppression for Radar via Deep Joint Sparse Recovery Network | |
| CN113848532A (en) | FMCW radar signal noise reduction system and method based on noise reduction model | |
| CN116520256B (en) | Airborne early warning radar interference recognition method and device based on deep learning | |
| Volosyuk et al. | Optimal radar cross section estimation in synthetic aperture radar | |
| Gulum | Autonomous non-linear classification of LPI radar signal modulations | |
| Wu et al. | Cv-sagan: Complex-valued self-attention gan on radar clutter suppression and target detection | |
| CN115061094B (en) | Radar target recognition method based on neural network and SVM | |
| CN115267713B (en) | Intermittent sampling interference identification and suppression method based on semantic segmentation | |
| Tian et al. | Multiresolution jamming recognition with few-shot learning | |
| Shreyamsha Kumar et al. | Target identification using harmonic wavelet based ISAR imaging |
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 |