CN114866311A - Radio frequency fingerprint extraction method based on time sequence representation - Google Patents
Radio frequency fingerprint extraction method based on time sequence representation Download PDFInfo
- Publication number
- CN114866311A CN114866311A CN202210464767.4A CN202210464767A CN114866311A CN 114866311 A CN114866311 A CN 114866311A CN 202210464767 A CN202210464767 A CN 202210464767A CN 114866311 A CN114866311 A CN 114866311A
- Authority
- CN
- China
- Prior art keywords
- radio frequency
- frequency fingerprint
- signal
- time sequence
- extraction method
- 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.)
- Granted
Links
- 238000000605 extraction Methods 0.000 title claims abstract description 26
- 238000000034 method Methods 0.000 claims abstract description 24
- 239000011159 matrix material Substances 0.000 claims abstract description 12
- 238000011084 recovery Methods 0.000 claims abstract description 10
- 239000013598 vector Substances 0.000 claims abstract description 10
- 238000009499 grossing Methods 0.000 claims abstract description 7
- 238000012545 processing Methods 0.000 claims abstract description 6
- 230000002776 aggregation Effects 0.000 claims abstract description 5
- 238000004220 aggregation Methods 0.000 claims abstract description 5
- 238000009877 rendering Methods 0.000 claims abstract description 4
- 230000011218 segmentation Effects 0.000 claims abstract description 4
- 238000012512 characterization method Methods 0.000 claims description 8
- 238000005070 sampling Methods 0.000 claims description 8
- 238000010606 normalization Methods 0.000 claims description 6
- 230000008569 process Effects 0.000 claims description 5
- 238000004364 calculation method Methods 0.000 claims description 4
- 238000006243 chemical reaction Methods 0.000 claims description 3
- 150000001875 compounds Chemical class 0.000 claims description 3
- 238000005520 cutting process Methods 0.000 claims description 3
- 238000001514 detection method Methods 0.000 claims description 3
- 238000007781 pre-processing Methods 0.000 claims description 3
- 238000005065 mining Methods 0.000 abstract description 2
- 238000013527 convolutional neural network Methods 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 239000002131 composite material Substances 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000005284 excitation Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000009432 framing Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000011176 pooling Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 238000010200 validation analysis Methods 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L63/00—Network architectures or network communication protocols for network security
- H04L63/08—Network architectures or network communication protocols for network security for authentication of entities
- H04L63/0876—Network architectures or network communication protocols for network security for authentication of entities based on the identity of the terminal or configuration, e.g. MAC address, hardware or software configuration or device fingerprint
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B17/00—Monitoring; Testing
- H04B17/30—Monitoring; Testing of propagation channels
- H04B17/309—Measuring or estimating channel quality parameters
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B17/00—Monitoring; Testing
- H04B17/30—Monitoring; Testing of propagation channels
- H04B17/309—Measuring or estimating channel quality parameters
- H04B17/336—Signal-to-interference ratio [SIR] or carrier-to-interference ratio [CIR]
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W12/00—Security arrangements; Authentication; Protecting privacy or anonymity
- H04W12/06—Authentication
-
- 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)
- Signal Processing (AREA)
- Computer Security & Cryptography (AREA)
- Quality & Reliability (AREA)
- Physics & Mathematics (AREA)
- Electromagnetism (AREA)
- Power Engineering (AREA)
- Computer Hardware Design (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Collating Specific Patterns (AREA)
Abstract
The invention discloses a radio frequency fingerprint extraction method based on time sequence representation, which comprises the following steps: collecting and processing an output signal of the wireless device; the scaled numerical value is encoded into cosine of a vector included angle, a corresponding timestamp is encoded into a radius, and a real number sequence in a Cartesian coordinate system is converted into a sequence in a polar coordinate system; calculating a class gram matrix corresponding to the sequence according to the inner product definition to serve as a triangular item of the radio frequency fingerprint; taking the zoomed numerical value as a main diagonal line to obtain a corresponding diagonal matrix as a recovery item of the radio frequency fingerprint; calculating the sum of the triangular item and the recovery item, and rendering the sum into an image according to the numerical value; and smoothing the image by using a segmentation aggregation approximation method, and taking the smoothed image as the radio frequency fingerprint of the wireless equipment. The method overcomes the problems of lack of time sequence information mining and large resource consumption in the traditional radio frequency fingerprint extraction algorithm, is safe and effective, and can be used as a radio frequency fingerprint extraction method of various wireless devices.
Description
Technical Field
The invention relates to the technical field of information security, in particular to a radio frequency fingerprint extraction method based on time sequence representation.
Background
The internet of things is combined with big data and cloud computing, and various industries are being deeply remodeled, and meanwhile, more serious challenges are brought to system safety. The openness of wireless networks makes them more vulnerable to attacks than wired networks. Thus, authentication of the wireless device is critical to maintaining system security. Physical layer authentication has become a powerful solution in wireless device access due to advantages of low cost and low delay, wherein radio frequency fingerprint is one of the most core technologies in the physical layer authentication process. As an inherent property of the device circuit hardware, radio frequency fingerprints are difficult to forge, as are biometric fingerprints, which are unique identifiers of devices. In addition, the radio frequency fingerprint has universality, short-time invariance, independence and robustness, and these advantages make the radio frequency fingerprint attract extensive attention and research.
However, most of the existing radio frequency fingerprint extraction schemes regard the collected signals as static data streams with independent data points, and ignore the time sequence relation of the signals as time sequences, which results in insufficient performance of subsequent identification and authentication. In addition, the existing radio frequency fingerprint extraction is designed under an ideal data set, namely, the complete preamble is often used for fingerprint extraction. The conditions of incomplete acquisition, signal loss and the like which are possibly caused in the acquisition process in a real scene are considered, the method is often insufficient in practical universality, and the problems of high calculation cost and the like are caused by the fingerprint extraction based on the complete lead code. Therefore, an efficient, lightweight, and highly practical method for extracting a radio frequency fingerprint is urgently needed.
Disclosure of Invention
The invention aims to: in order to overcome the defects of the existing fingerprint extraction scheme, the invention provides a radio frequency fingerprint extraction method based on time sequence representation, which is used for solving the problems of lack of time sequence information mining, insufficient practicability and large resource consumption in the existing extraction method.
The technical scheme is as follows: after the signal to be identified is obtained, the signal to be identified is preprocessed, then one-dimensional data is processed into a two-dimensional image, and finally the image obtained through smoothing operation is used as a radio frequency fingerprint. The radio frequency fingerprint extraction method based on the time sequence representation comprises the following steps:
(1) collecting and processing an output signal of the wireless device;
(2) the scaled numerical value is encoded into cosine of a vector included angle, a corresponding timestamp is encoded into a radius, and a real number sequence in a Cartesian coordinate system is converted into a sequence in a polar coordinate system;
(3) calculating a class gram matrix corresponding to the sequence according to the inner product definition to serve as a triangular item of the radio frequency fingerprint;
(4) taking the zoomed numerical value as a main diagonal line to obtain a corresponding diagonal matrix as a recovery item of the radio frequency fingerprint;
(5) calculating the sum of the triangular item and the recovery item, and rendering the sum into an image according to the numerical value;
(6) and smoothing the image by using a segmentation aggregation approximation method, and taking the smoothed image as the radio frequency fingerprint of the wireless equipment.
Further, the step (1) specifically comprises:
(1.1) acquiring an output signal of the wireless equipment;
(1.2) preprocessing the acquired signals: and sequentially carrying out down-conversion, oversampling, signal detection and interception, energy normalization, estimation and compensation of signal frequency offset and phase offset and I/Q path signal extraction.
(1.3) carrying out minimum-maximum normalization on the preprocessed signals, and scaling the data to a [0,1] interval range. The min-max formula is:
wherein X represents the acquired wireless signal, X ═ X 1 ,x 2 ,…,x n }。
Further, the formula of the encoding process in step (2) is as follows:
in the formula, t i Is the time stamp, and N is the total number of sampling points of the signal acquired.
Further, the step (3) specifically comprises:
(3.1) defining the inner product calculation formula of any two vectors y and z as follows:
and (3.2) calculating a gram-like matrix of the preprocessed signal as a triangular term according to the new inner product definition. The formula is as follows:
wherein, I is a unit column vector,is the wireless signal obtained after the step (1), is thatThe transpose of (a) is performed,
further, the formula of the recovery term in step (4) is specifically as follows:
in the formula (I), the compound is shown in the specification, is the wireless signal obtained after the step (1),
further, the step (6) specifically comprises:
(6.1) setting window proportion m (0< m ≦ 1), wherein m is the proportion of the window size and the total sequence length. The size of each window is mN;
(6.2) cutting the original sequence into a plurality of segments in a non-overlapping mode through the window, and calculating the average value of each segment to replace all values in the segment.
Has the advantages that: compared with the prior art, the invention has the following remarkable advantages: by the method, the radio frequency fingerprint of the equipment can be extracted by adopting few lead code information, and the time sequence information of the signal is excavated through the time sequence representation of the gram-like matrix, so that the effective and light radio frequency fingerprint is provided for the authentication and identification of the wireless equipment in a real scene, and the method has high practical value. The method can be obtained through simulation and experiments, and the identification performance of the wireless equipment can be greatly improved by using the method.
Drawings
Fig. 1 is a schematic flow chart of an embodiment of a radio frequency fingerprint extraction method based on time series characterization according to the present invention.
Detailed Description
The embodiment provides a radio frequency fingerprint extraction method based on time sequence characterization, as shown in fig. 1, including the following steps:
(1) the output signal of the wireless device is collected and processed. The method comprises the following steps:
and (1.1) acquiring an output signal of the wireless equipment.
When the output signal of the wireless equipment is collected, the signal is collected through connecting the direct coaxial line with an attenuator, or the signal is collected in a wireless receiving environment with a short distance, a visible distance and a signal-to-noise ratio higher than a preset value. In the embodiment, 15 ZigBee wireless transmitting modules are selected as target wireless equipment and numbered according to 1-15. And (3) acquiring a line-of-sight transmission signal at a short distance by adopting USRP equipment, wherein the signal-to-noise ratio of the acquired signal is 30 dB. In this embodiment, each ZigBee device acquires 45 frames in total, and 675 frames in total. By artificially injecting white Gaussian noise, the signal-to-noise ratios of the signals reach 10dB,15dB,20dB,25dB and 30dB respectively, and the total frame number is expanded to 3375 frames.
(1.2) preprocessing the output signal of each wireless device.
Wherein the pretreatment comprises: down-conversion, over-sampling, signal detection and interception, energy normalization, estimation and compensation of signal frequency offset and phase offset, and I/Q path signal extraction. The symbol rate of original ZigBee equipment is 1Mbps, directly down-converts the signal to a baseband signal during acquisition, the sampling rate is 10Mbps, after framing the received signal according to the change of sampling points, energy normalization is performed on each frame, and finally frequency offset and phase offset processing of the signal are performed (refer to patent 201510797097.8 for a specific method).
(1.3) selecting the I-path signal, carrying out minimum-maximum processing on the I-path signal, and zooming the data to the range of [0,1], wherein the formula is as follows:
wherein X represents the preprocessed I-channel signal, and X is { X ═ X 1 ,x 2 ,…,x n }。
(2) And coding the scaled numerical value into cosine of a vector included angle, coding a corresponding timestamp into a radius, and converting a real number sequence in a Cartesian coordinate system into a sequence in a polar coordinate system. The formula of the encoding process is:
in the formula, t i Is a time stamp, and N is a signal of signal acquisitionThe total number of sample points. In this embodiment, 320 sampling points from the second front to the back are taken, i.e., N equals 320. In a specific application, the total number of sampling points is determined according to the signal acquisition condition.
(3) And calculating a class gram matrix corresponding to the sequence according to the inner product definition to serve as a triangular item of the radio frequency fingerprint. The method comprises the following steps:
(3.1) defining the inner product calculation formula of any two vectors y and z as follows:
and (3.2) calculating a gram-like matrix of the preprocessed signal as a triangular term according to the new inner product definition. The formula is as follows:
wherein, I is a unit column vector,is the wireless signal obtained after the step (1), is thatThe transpose of (a) is performed,
(4) taking the I path signal as a main diagonal line to obtain a corresponding diagonal matrix as a recovery item of the radio frequency fingerprint, wherein the formula is as follows:
in the formula (I), the compound is shown in the specification, is the wireless signal obtained after the step (1),
(5) and calculating the sum of the triangle item and the recovery item, and rendering the sum into an image according to the numerical value. In this embodiment, after the step, the size of the image corresponding to each frame signal is 320 × 320.
(6) And smoothing the image by using a segmentation aggregation approximation method, and taking the smoothed image as the radio frequency fingerprint of the wireless equipment. The method specifically comprises the following steps:
(6.1) setting window proportion m (0< m ≦ 1), wherein m is the proportion of the window size and the total sequence length. The size of each window is mN;
(6.2) cutting the original sequence into a plurality of segments in a non-overlapping mode through the window, and calculating the average value of each segment to replace all values in the segment.
In this embodiment, since m is 0.03125, the size of each window is 10, and the image size after smoothing is 32 × 32. The window proportion is used as a hyper-parameter, and the specific size is determined according to the total number of sampling points and the subsequent classification performance.
After the above steps, the obtained data aggregate is 3375, according to 7: 1: a scale of 2 divides the data set into a training set, a validation set, and a test set. To demonstrate the effectiveness of the present invention, a convolutional neural network was used for high-dimensional feature extraction and classification. The structure of the convolutional neural network is shown in table 1:
TABLE 1
Network layer | Dimension of input | Convolution kernel/step size/zero padding | Step size of pooling | Output dimension | Excitation function |
Conv1 | 400×400×3 | 3×3×32/3/1 | - | 134×134×32 | BN+ReLU |
Pool1 | 134×134×32 | - | 2 | 67×67×32 | - |
Conv2 | 32×67×67 | 3×3×64/3/1 | - | 23×23×64 | BN+ReLU |
Pool2 | 23×23×64 | - | 2 | 11×11×64 | - |
Conv3 | 11×11×64 | 3×3×64/2/1 | - | 6×6×64 | BN+ReLU |
Pool3 | 6×6×64 | - | 2 | 3×3×64 | - |
Flatten | 3×3×64 | - | - | 576 | - |
FC1 | 576 | - | - | 64 | Dropout+ReLU |
FC2 | 64 | - | - | 15 | Softmax |
By the method, the accuracy of identifying and classifying the low-power-consumption equipment can be effectively improved. As shown in table 2, compared with the original signal after being directly preprocessed, the classification accuracy is significantly improved after the time sequence characterization is performed by using the method.
TABLE 2
Signal-to-noise ratio (dB) | 10 | 15 | 20 | 25 | 30 |
Original signal | 62.9629 | 71.8518 | 75.5556 | 77.0370 | 80 |
Method for producing a composite material | 92.5925 | 96.2962 | 97.7778 | 99.2593 | 99.2593 |
While the invention has been described in connection with what is presently considered to be the most practical and preferred embodiment, it is to be understood that the invention is not to be limited to the disclosed embodiment, but on the contrary, is intended to cover various modifications and equivalent arrangements included within the spirit and scope of the appended claims.
Claims (6)
1. A radio frequency fingerprint extraction method based on time sequence representation is characterized by comprising the following steps:
(1) collecting and processing an output signal of the wireless device;
(2) the scaled numerical value is encoded into cosine of a vector included angle, a corresponding timestamp is encoded into a radius, and a real number sequence in a Cartesian coordinate system is converted into a sequence in a polar coordinate system;
(3) calculating a class gram matrix corresponding to the sequence according to the inner product definition to serve as a triangular item of the radio frequency fingerprint;
(4) taking the zoomed numerical value as a main diagonal line to obtain a corresponding diagonal matrix as a recovery item of the radio frequency fingerprint;
(5) calculating the sum of the triangular item and the recovery item, and rendering the sum into an image according to the numerical value;
(6) and smoothing the image by using a segmentation aggregation approximation method, and taking the smoothed image as the radio frequency fingerprint of the wireless equipment.
2. The radio frequency fingerprint extraction method based on time sequence characterization according to claim 1, wherein: the step (1) specifically comprises:
(1.1) acquiring an output signal of the wireless equipment;
(1.2) preprocessing the acquired signals: sequentially carrying out down-conversion, oversampling, signal detection and interception, energy normalization, estimation and compensation of signal frequency offset and phase offset and I/Q path signal extraction;
(1.3) carrying out minimum-maximum normalization on the preprocessed signals, and scaling the data to a range of [0,1], wherein the minimum-maximum formula is as follows:
wherein X represents the acquired wireless signal, X ═ X { X } 1 ,x 2 ,…,x n }。
4. The radio frequency fingerprint extraction method based on time sequence characterization according to claim 1, wherein: the step (3) specifically comprises:
(3.1) defining the inner product calculation formula of any two vectors y and z as follows:
(3.2) according to the definition of the new inner product, calculating a gram-like matrix of the preprocessed signal as a triangular term, wherein the formula is as follows:
5. the radio frequency fingerprint extraction method based on time sequence characterization according to claim 1, wherein: the formula of the recovery item in the step (4) is specifically as follows:
6. the radio frequency fingerprint extraction method based on time sequence characterization according to claim 1, wherein: in the step (6), the smoothing processing is performed on the image by using a piecewise aggregation approximation method, which specifically includes:
(6.1) setting window proportion m, wherein m is more than 0 and less than or equal to 1, and m is the proportion of the window size to the total sequence length, so that the size of each window is mN;
(6.2) cutting the original sequence into a plurality of sections in a non-overlapping mode through the window, and calculating the average value of each section to replace all values in the section;
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210464767.4A CN114866311B (en) | 2022-04-29 | 2022-04-29 | Radio frequency fingerprint extraction method based on time sequence characterization |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210464767.4A CN114866311B (en) | 2022-04-29 | 2022-04-29 | Radio frequency fingerprint extraction method based on time sequence characterization |
Publications (2)
Publication Number | Publication Date |
---|---|
CN114866311A true CN114866311A (en) | 2022-08-05 |
CN114866311B CN114866311B (en) | 2024-03-22 |
Family
ID=82636125
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210464767.4A Active CN114866311B (en) | 2022-04-29 | 2022-04-29 | Radio frequency fingerprint extraction method based on time sequence characterization |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114866311B (en) |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105357014A (en) * | 2015-11-25 | 2016-02-24 | 东南大学 | Wireless equipment radio frequency fingerprint feature extraction method based on differential constellation track diagram |
CN105631472A (en) * | 2015-12-24 | 2016-06-01 | 东南大学 | Wireless device identity identification method based on constellation locus diagram |
CN110380989A (en) * | 2019-07-26 | 2019-10-25 | 东南大学 | The polytypic internet of things equipment recognition methods of network flow fingerprint characteristic two-stage |
US20200036710A1 (en) * | 2018-07-25 | 2020-01-30 | Honda Motor Co., Ltd. | Method and system for encryption using a radio frequency fingerprint |
CN114239657A (en) * | 2021-12-18 | 2022-03-25 | 中国科学技术大学 | Time sequence signal identification method based on complex value interference neural network |
-
2022
- 2022-04-29 CN CN202210464767.4A patent/CN114866311B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105357014A (en) * | 2015-11-25 | 2016-02-24 | 东南大学 | Wireless equipment radio frequency fingerprint feature extraction method based on differential constellation track diagram |
CN105631472A (en) * | 2015-12-24 | 2016-06-01 | 东南大学 | Wireless device identity identification method based on constellation locus diagram |
US20200036710A1 (en) * | 2018-07-25 | 2020-01-30 | Honda Motor Co., Ltd. | Method and system for encryption using a radio frequency fingerprint |
CN110380989A (en) * | 2019-07-26 | 2019-10-25 | 东南大学 | The polytypic internet of things equipment recognition methods of network flow fingerprint characteristic two-stage |
CN114239657A (en) * | 2021-12-18 | 2022-03-25 | 中国科学技术大学 | Time sequence signal identification method based on complex value interference neural network |
Non-Patent Citations (3)
Title |
---|
季澈;彭林宁;胡爱群;王栋: "基于射频信号特征的Airmax设备指纹提取方法", 数据采集与处理, no. 02 * |
王栋;胡爱群;王炎: "多径传输环境下的射频指纹估计方法", 密码学报, no. 02 * |
黄少驰;朱晓蕾: "一种基于射频指纹的通信个体识别方法", 航天电子对抗, no. 01 * |
Also Published As
Publication number | Publication date |
---|---|
CN114866311B (en) | 2024-03-22 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Babu et al. | Statistical features based optimized technique for copy move forgery detection | |
CN109035149B (en) | License plate image motion blur removing method based on deep learning | |
CN109495214B (en) | Channel coding type identification method based on one-dimensional inclusion structure | |
CN110197209B (en) | Radiation source identification method based on multi-feature fusion | |
CN113014524B (en) | Digital signal modulation identification method based on deep learning | |
CN110751612A (en) | Single image rain removing method of multi-channel multi-scale convolution neural network | |
CN101763514B (en) | Image segmentation method based on characteristic importance sorting spectral clustering | |
CN111339924B (en) | Polarized SAR image classification method based on superpixel and full convolution network | |
Yu et al. | Robust median filtering forensics by CNN-based multiple residuals learning | |
CN111027614B (en) | Noise enhanced radio frequency fingerprint identification method and device | |
CN110532825A (en) | A kind of bar code identifying device and method based on artificial intelligence target detection | |
CN111624570A (en) | Radar target identification method based on two-dimensional convolutional neural network | |
CN1564600A (en) | Detection method of moving object under dynamic scene | |
CN104537381A (en) | Blurred image identification method based on blurred invariant feature | |
CN110135390B (en) | Radiation source individual identification method based on main signal suppression | |
CN114866311A (en) | Radio frequency fingerprint extraction method based on time sequence representation | |
CN110536299B (en) | Data packet validity confirmation method based on edge calculation and discrete random convolution | |
CN115225731B (en) | Online protocol identification method based on hybrid neural network | |
CN114584441A (en) | Digital signal modulation identification method based on deep learning | |
CN115510905A (en) | Multitask learning method for blind identification of channel coding | |
CN115147895A (en) | Face counterfeit discrimination method and device and computer program product | |
Weng et al. | Message structure aided attentional convolution network for RF device fingerprinting | |
CN114330443A (en) | Indoor crowd counting method based on channel state information | |
CN101937511B (en) | Rapid image matching method based on stochastic parallel optimization algorithm | |
CN107395540A (en) | Modulation Signals Recognition method based on LMD approximate entropies, high accumulation accumulated amount and SVM |
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 |