CN112911597A - Internet of things physical layer multilevel feature extraction method based on radio frequency signal fine portrait - Google Patents
Internet of things physical layer multilevel feature extraction method based on radio frequency signal fine portrait Download PDFInfo
- Publication number
- CN112911597A CN112911597A CN202110330439.0A CN202110330439A CN112911597A CN 112911597 A CN112911597 A CN 112911597A CN 202110330439 A CN202110330439 A CN 202110330439A CN 112911597 A CN112911597 A CN 112911597A
- Authority
- CN
- China
- Prior art keywords
- signal
- radio frequency
- signals
- portrait
- internet
- 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
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W12/00—Security arrangements; Authentication; Protecting privacy or anonymity
- H04W12/06—Authentication
-
- 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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W24/00—Supervisory, monitoring or testing arrangements
- H04W24/08—Testing, supervising or monitoring using real traffic
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Signal Processing (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computer Networks & Wireless Communication (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Computer Security & Cryptography (AREA)
- Medical Informatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses an Internet of things physical layer multilevel feature extraction method based on a radio frequency signal fine portrait, which comprises the following steps: performing multi-level feature extraction and portrait representation of a primary signal, and depicting fine features of individual signals of different types, models and batches from multiple levels to form a primary portrait; for individual equipment with secondary signal characteristics, intelligently fusing multi-attribute characteristics of signals by utilizing deep learning according to the classifiable degree of the signals to be identified, more comprehensively describing multi-dimensional gene fusion characteristics of the signals and forming a secondary portrait; according to the method of the equipotential planet map based on the fusion characteristics, the multi-dimensional vector characteristics are converted into multi-dimensional color images, the intelligent perception method of the deep learning algorithm is combined, the multi-reliability intelligent fusion is carried out on different image characteristics, the information content of characteristic difference among radio frequency signal samples is increased, and the three-level fine image is formed. The invention can extract the fine portrait characteristics of signal with multilevel, multidimensional and multi-granularity.
Description
Technical Field
The invention belongs to the technical field of communication, and particularly relates to a method for extracting multilevel characteristics of a physical layer of an Internet of things based on a radio frequency signal fine portrait.
Background
The feature extraction of the physical layer signal of the Internet of things mainly provides a reliable feature database for the security certification of the Internet of things. The traditional authentication mechanism is realized in an application layer, a numerical result which is difficult to counterfeit by a third party is generated by utilizing a cryptographic algorithm, but the mechanism has risks of protocol security loopholes, secret key leakage and the like. The internet of things sensing layer terminal equipment has the characteristics of diversification, intellectualization, complexity and huge quantity, and although the traditional authentication mechanism can ensure information safety to a certain extent, the internet of things sensing layer terminal equipment is not suitable for processing large-scale networks and mass data brought by the large-scale networks, and the information safety requirements of the internet of things are difficult to meet. The field security manufacturer, cimagnek, combines with the largest chip provider in the world, texas instruments, and the cryptographic service provider, wolfSSL, to combine authentication technology, encryption technology, and embedded technology together to provide reliable information encryption and identity authentication for internet of things devices, however, this device protection method based on embedded technology requires higher cost. For the identity authentication of terminal equipment of the power transmission and transformation internet of things and the power distribution internet of things, a barcode and an RFID technology based on identity ID identification are commonly used. The RFID technology essentially performs device identity authentication by using intentionally modulated information carried by electromagnetic waves after demodulation, but security and privacy threats of the RFID technology relate to problems such as eavesdropping, counterfeiting, and tag cloning [1 ]. Therefore, the research of an identity identification authentication method [2,3] for access and control of the sensing layer terminal equipment with low error rate, high efficiency and low cost is a key for ensuring the stable operation of the Internet of things.
The essence of the physical layer access authentication of the internet of things is a radio frequency fingerprint identification technology. The existing radio frequency fingerprint identification technology can be divided into a channel-based fingerprint identification technology and a transmission signal-based fingerprint identification technology according to different physical layer resources. The fingerprint identification technology based on the channel characteristics aims to utilize the unique position information of the equipment as the identity detection indexes of different users in different scenes, and is generally applied to indoor positioning of the equipment of the Internet of things. Common channel characteristics are Radio Signal Strength (RSS), Channel State Information (CSI), and Channel Frequency Response (CFR). The fingerprint recognition technology based on the transmission signal is classified into a fingerprint recognition technology based on a transient signal and a fingerprint recognition technology based on a steady-state signal. The transient signal does not contain any data information, only reflects the hardware characteristics of a transmitter, has independence, and the radio frequency fingerprint is extracted from the transient signal at first, such as the duration, fractal dimension characteristics, spectral characteristics, time domain envelope, wavelet coefficient and the like of the transient signal. Because the transient signal has short duration, is difficult to capture, is sensitive to the detection and the positioning of the mutation point, and limits the application of the transient signal in the actual environment. The steady-state signal is a signal when a transmitter is in a stable working state, has long duration and is more easily obtained and can be completed by using a cheap receiver, but radio frequency fingerprints existing in the steady-state signal are less easily extracted, such as frequency offset, Holder coefficient characteristics, entropy characteristics and the like. With the development of the radio frequency fingerprint identification technology, students gradually reduce the requirements on detection and extraction of signals to be identified from a preamble sequence utilizing a transient signal to a preamble sequence utilizing a steady-state signal to a transmission data segment utilizing the steady-state signal.
[1]C.Bertoncini,K.Rudd,B.Nousain,et al.Wavelet fingerprinting of radio-frequency identification(RFID)tags[J].IEEE Transactions on Industrial Electronics,2012,59(12):4843-4850.
[2]Dajiang Chen,Ning Zhang,Zhen Qin,et al.S2M:A lightweight acoustic fingerprints-based wireless device authentication protocol[J].IEEE Internet of Things Journal,2017,4(1):88-100.
[3] Great tiger, Chenxiang, Linyun, etc. research status and trend of radio frequency fingerprint identification [ J ] electric wave science, 2019 DOI 10.13443/j.cjors.2019070501
Disclosure of Invention
The invention aims to provide a method for extracting multilevel characteristics of a physical layer of the Internet of things based on a radio frequency signal fine portrait so as to solve the problems in the prior art.
The technical scheme of the invention is that an Internet of things physical layer multilevel feature extraction method based on a radio frequency signal fine portrait comprises the following steps:
a multilevel feature extraction method for an Internet of things physical layer based on a radio frequency signal fine portrait comprises the following steps:
(S1) performing a first-level signal multi-level feature extraction and an image representation; selecting slice scales according to the distribution characteristics of radio frequency signals to be identified, dividing the signals into multidimensional slices, and extracting the characteristics of each slice, so that the fine characteristics of individual signals of different types, models and batches can be more finely described from multiple levels to form a primary image;
(S2) performing two-level signal multi-dimensional feature intelligent fusion and portrait representation; for individual equipment with secondary signal characteristics, carrying out intelligent fusion on the multidimensional characteristics of the signals by utilizing deep learning according to the classifiable degree of the signals to be identified to obtain the multidimensional gene fusion characteristics for describing the signals, and forming a secondary portrait;
(S3) performing three-level signal multi-confidence image fusion and representation; according to the method of the equipotential planet map based on the fusion characteristics, the multidimensional characteristics of the signals are converted into multidimensional color images, the intelligent perception method of the deep learning algorithm is combined, the multi-reliability intelligent fusion is carried out on different image characteristics, the information content of characteristic difference among radio frequency signal samples is increased, and the three-level fine image is formed.
The invention is further improved in that: the one-level signal multi-level feature extraction in step S1 is to cut a sample signal with a data length of n points into equal slices.
The invention is further improved in that: in step S2, the multi-dimension means that each dimension characteristic of the signal is extracted from the time domain, the wavelet domain, the fractal domain, the entropy domain, the FRFT domain, and the FFT domain.
The invention is further improved in that: in step S3, the multi-granularity is to convert the multi-dimensional vector features into color images, perform intelligent fusion on different image features, and increase the information amount of feature differences between the radio frequency signal samples to distinguish the weights of different features.
The invention is further improved in that: in step S3, converting the multi-dimensional features of the signal into a multi-dimensional color image, comprising the steps of: and forming a two-dimensional image by the multi-dimensional features obtained in the step S2, and coloring the two-dimensional image by using an equipotential planet map.
The invention has the beneficial effects that: the traditional authentication mechanism is realized in an application layer, a numerical result which is difficult to counterfeit by a third party is generated by utilizing a cryptographic algorithm, but the mechanism has risks of protocol security loopholes, secret key leakage and the like. The individual identification of the radiation source based on the fingerprint of the radio frequency signal is based on the difference of hardware of a physical layer of equipment and does not change theoretically along with the change of a transmission signal. The invention uses the characteristic of the radio frequency signal to extract the fine portrait characteristics of multiple layers, multiple dimensions and multiple granularities of the signal, thereby providing a reliable characteristic database for the accurate identification and authentication of the physical layer signal.
Drawings
FIG. 1 is a flow chart of a method for extracting multilevel features of a physical layer of the Internet of things based on a radio frequency signal fine portrait.
Detailed Description
Because the individual characteristic information of the radiation source of the Internet of things is attached to the original signal, the individual characteristic difference is small, the digital measurement is directly carried out by using a multi-characteristic fusion method, and the accurate identification of individual equipment is difficult to realize. Defining terminal equipment of the internet of things with different models and different manufacturers or larger differences among the same model and different manufacturers in the internet of things equipment as primary signal equipment which is easy to identify; defining the Internet of things terminal equipment of the same manufacturer, the same model and different batches as second-level signal equipment, wherein the second-level signal equipment is relatively difficult to identify; the internet of things terminal equipment with the same manufacturer, the same model and the same batch with extremely high similarity is defined to be three-level signal equipment, and the classification of the internet of things terminal equipment is difficult to realize by a general feature extraction and classification algorithm and is also the content of key research of the invention. The invention introduces a user portrait technology for describing user behavior habits in the marketing field into the field of individual identification of radiation sources of the physical layer of the Internet of things, and focuses on researching a feature extraction method of equipment with three-level signal characteristics on the basis of accurately identifying terminal equipment emitting primary and secondary signals.
As shown in fig. 1, an embodiment of the present invention includes the steps of:
(S1) performing a first-level signal multi-level feature extraction and an image representation; according to the distribution characteristics of the radio frequency signals to be identified, slice scales are selected, the signals are divided into multi-dimensional slices, feature extraction is carried out on each slice, and the fine features of individual signals of different types, models and batches are more finely described from multiple levels to form a primary portrait.
The one-level signal multi-level feature extraction refers to cutting a sampling signal with the data length of n points into equal slices. For example, a sampled signal with a data length of 2000 points is cut every 200 points and divided into 10 slices, and data of each segment is allowed to be partially overlapped;
in the process of selecting the slice scale, different scales contain different numbers of discrete points, and the number is determined according to the distribution characteristics of the signal points.
(S2) performing two-level signal multi-dimensional feature intelligent fusion and portrait representation; for individual equipment with secondary signal characteristics, according to the classifiable degree of the signal to be identified, the multidimensional characteristics of the signal are intelligently fused by utilizing deep learning, wherein the multidimensional characteristics refer to the characteristics of each dimension of the signal extracted from a time domain, a wavelet domain, a fractal domain, an entropy domain, an FRFT domain and an FFT domain, and the characteristics are specificThe method comprises the following steps of (1) including power spectrum characteristics, representing the distribution condition of signal frequencies in a frequency domain; the dual-spectrum characteristic is used for detecting and representing nonlinearity in the signals and better representing subtle differences among different signals; the displacement entropy and signal mutation detection method can conveniently and accurately position the time when a system mutates and has amplification effect on the tiny change of signals; instantaneous amplitude envelope characteristics are subjected to Hilbert transform to obtain analytic representation of the signal, so that the instantaneous frequency and the phase of the signal are calculated; a Wigner-Ville distribution (WVD) feature that describes the time-frequency energy distribution of the signal; RF-DNA fingerprinting, calculating 3 statistical characteristics of signal-variance (sigma)2) Deviation (gamma) and kurtosis (k) to obtain the multi-dimensional gene fusion characteristics of the description signals, and form a secondary portrait.
(S3) performing three-level signal multi-confidence image fusion and representation; according to the method of the equipotential planet map based on the fusion characteristics, the multidimensional characteristics of the signals are converted into multidimensional color images, the intelligent perception method of the deep learning algorithm is combined, the multi-reliability intelligent fusion is carried out on different image characteristics, the information content of characteristic difference among radio frequency signal samples is increased, and the three-level fine image is formed.
In step S3, the multi-granularity is to convert the multi-dimensional vector features into color images, perform intelligent fusion on different image features, and increase the information amount of feature differences between the radio frequency signal samples to distinguish the weights of different features.
In step S3, converting the multi-dimensional features of the signal into a multi-dimensional color image, comprising the steps of: and forming a two-dimensional image by the multi-dimensional features obtained in the step S2, and coloring the two-dimensional image by using an equipotential planet map.
The traditional authentication mechanism is realized in an application layer, a numerical result which is difficult to counterfeit by a third party is generated by utilizing a cryptographic algorithm, but the mechanism has risks of protocol security loopholes, secret key leakage and the like. The individual identification of the radiation source based on the fingerprint of the radio frequency signal is based on the difference of hardware of a physical layer of equipment and does not change theoretically along with the change of a transmission signal. The invention uses the characteristic of the radio frequency signal to extract the fine portrait characteristics of multiple layers, multiple dimensions and multiple granularities of the signal, thereby providing a reliable characteristic database for the accurate identification and authentication of the physical layer signal.
The research on the physical layer security authentication technology is still in a preliminary stage nowadays, rich physical layer resources are not fully utilized, and a large research space is still provided. The radio frequency fingerprint identification is a non-password authentication method based on equipment physical layer hardware, does not need to consume additional computing resources, does not need to embed additional hardware, and is a very potential technology for constructing a low-cost, simpler and safer identification and authentication system. Therefore, the key for ensuring the steady operation of the ubiquitous power internet of things is to research the identity recognition authentication method for the access and control of the sensing layer terminal equipment with low error rate, high efficiency and low cost.
The foregoing detailed description of the preferred embodiments of the invention has been presented. It should be understood that numerous modifications and variations could be devised by those skilled in the art in light of the present teachings without departing from the inventive concepts. Therefore, the technical solutions available to those skilled in the art through logic analysis, reasoning and limited experiments based on the prior art according to the concept of the present invention should be within the scope of protection defined by the claims.
Claims (5)
1. A multilevel feature extraction method for an Internet of things physical layer based on a radio frequency signal fine portrait comprises the following steps:
(S1) performing a first-level signal multi-level feature extraction and an image representation; selecting slice scales according to the distribution characteristics of radio frequency signals to be identified, dividing the signals into multidimensional slices, and extracting the characteristics of each slice, so that the fine characteristics of individual signals of different types, models and batches can be more finely described from multiple levels to form a primary image;
(S2) performing two-level signal multi-dimensional feature intelligent fusion and portrait representation; for individual equipment with secondary signal characteristics, carrying out intelligent fusion on the multidimensional characteristics of the signals by utilizing deep learning according to the classifiable degree of the signals to be identified to obtain the multidimensional gene fusion characteristics for describing the signals, and forming a secondary portrait;
(S3) performing three-level signal multi-confidence image fusion and representation; according to the method of the equipotential planet map based on the fusion characteristics, the multidimensional characteristics of the signals are converted into multidimensional color images, the intelligent perception method of the deep learning algorithm is combined, the multi-reliability intelligent fusion is carried out on different image characteristics, the information content of characteristic difference among radio frequency signal samples is increased, and the three-level fine image is formed.
2. The method for extracting the multilevel features of the physical layer of the internet of things based on the radio frequency signal fine portrait according to claim 1, wherein: the one-level signal multi-level feature extraction in step S1 is to cut a sample signal with a data length of n points into equal slices.
3. The method for extracting the multilevel features of the physical layer of the internet of things based on the radio frequency signal fine portrait according to claim 2, wherein: in step S2, the multi-dimension means that each dimension characteristic of the signal is extracted from the time domain, the wavelet domain, the fractal domain, the entropy domain, the FRFT domain, and the FFT domain.
4. The method for extracting the multilevel features of the physical layer of the internet of things based on the radio frequency signal fine sketch as claimed in claim 3, wherein: in step S3, the multi-granularity is to convert the multi-dimensional vector features into color images, perform intelligent fusion on different image features, and increase the information amount of feature differences between the radio frequency signal samples to distinguish the weights of different features.
5. The method for extracting the multilevel features of the physical layer of the internet of things based on the radio frequency signal fine sketch as claimed in claim 4, wherein: in step S3, converting the multi-dimensional features of the signal into a multi-dimensional color image, comprising the steps of: and forming a two-dimensional image by the multi-dimensional features obtained in the step S2, and coloring the two-dimensional image by using an equipotential planet map.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110330439.0A CN112911597B (en) | 2021-03-24 | 2021-03-24 | Internet of things physical layer multilevel feature extraction method based on radio frequency signal fine portrait |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110330439.0A CN112911597B (en) | 2021-03-24 | 2021-03-24 | Internet of things physical layer multilevel feature extraction method based on radio frequency signal fine portrait |
Publications (2)
Publication Number | Publication Date |
---|---|
CN112911597A true CN112911597A (en) | 2021-06-04 |
CN112911597B CN112911597B (en) | 2022-11-01 |
Family
ID=76109124
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110330439.0A Active CN112911597B (en) | 2021-03-24 | 2021-03-24 | Internet of things physical layer multilevel feature extraction method based on radio frequency signal fine portrait |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112911597B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113612751A (en) * | 2021-07-28 | 2021-11-05 | 深圳供电局有限公司 | Power distribution network power line carrier communication system access security detection method |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108809355A (en) * | 2018-07-04 | 2018-11-13 | 南京东科优信网络安全技术研究院有限公司 | A method of in low signal-to-noise ratio extraction equipment physical fingerprint feature |
CN108960138A (en) * | 2018-07-03 | 2018-12-07 | 南方电网科学研究院有限责任公司 | Equipment authentication feature identification method based on convolutional neural network |
CN109919015A (en) * | 2019-01-28 | 2019-06-21 | 东南大学 | A kind of radio-frequency fingerprint extraction and recognition methods based on more sampling convolutional neural networks |
CN110647951A (en) * | 2019-11-27 | 2020-01-03 | 南京邮电大学 | Wireless radio frequency equipment identity recognition method and system based on machine learning algorithm |
CN111563227A (en) * | 2020-04-30 | 2020-08-21 | 中国人民解放军军事科学院评估论证研究中心 | Fingerprint characteristic parameter extraction method for radiation source signal and radiation source identification |
CN112398553A (en) * | 2020-11-03 | 2021-02-23 | 上海电机学院 | Communication radiation source individual identification method based on differential equipotential sphere diagram |
-
2021
- 2021-03-24 CN CN202110330439.0A patent/CN112911597B/en active Active
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108960138A (en) * | 2018-07-03 | 2018-12-07 | 南方电网科学研究院有限责任公司 | Equipment authentication feature identification method based on convolutional neural network |
CN108809355A (en) * | 2018-07-04 | 2018-11-13 | 南京东科优信网络安全技术研究院有限公司 | A method of in low signal-to-noise ratio extraction equipment physical fingerprint feature |
CN109919015A (en) * | 2019-01-28 | 2019-06-21 | 东南大学 | A kind of radio-frequency fingerprint extraction and recognition methods based on more sampling convolutional neural networks |
CN110647951A (en) * | 2019-11-27 | 2020-01-03 | 南京邮电大学 | Wireless radio frequency equipment identity recognition method and system based on machine learning algorithm |
CN111563227A (en) * | 2020-04-30 | 2020-08-21 | 中国人民解放军军事科学院评估论证研究中心 | Fingerprint characteristic parameter extraction method for radiation source signal and radiation source identification |
CN112398553A (en) * | 2020-11-03 | 2021-02-23 | 上海电机学院 | Communication radiation source individual identification method based on differential equipotential sphere diagram |
Non-Patent Citations (4)
Title |
---|
XU JINYONG, ZHAO HANGSHENG: "Detection to Primary User Based on Radio Frequency Fingerprint in", 《IEEE》 * |
曾勇虎等: "射频指纹识别的研究现状及趋势", 《电波科学学报》 * |
李靖超,陈志敏: "基于多重分形维数的改进信号特征提取算法", 《上海电机学院学报》 * |
郑雅文,任佳鑫,刘铭: "基于深度神经网络的设备身份识别研究", 《研究与开发》 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113612751A (en) * | 2021-07-28 | 2021-11-05 | 深圳供电局有限公司 | Power distribution network power line carrier communication system access security detection method |
Also Published As
Publication number | Publication date |
---|---|
CN112911597B (en) | 2022-11-01 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Liu et al. | Zero-bias deep learning for accurate identification of Internet-of-Things (IoT) devices | |
Peng et al. | Deep learning based RF fingerprint identification using differential constellation trace figure | |
Shi et al. | Particle swarm optimization-based deep neural network for digital modulation recognition | |
Han et al. | GenePrint: Generic and accurate physical-layer identification for UHF RFID tags | |
Zhou et al. | A robust radio-frequency fingerprint extraction scheme for practical device recognition | |
Deng et al. | Radio frequency fingerprint extraction based on multidimension permutation entropy | |
Tu et al. | Research on the Internet of Things device recognition based on RF-fingerprinting | |
WO2021103206A1 (en) | Machine learning algorithm-based wireless radio frequency device identity recognition method and system | |
Wang et al. | Automatic modulation classification based on joint feature map and convolutional neural network | |
Li et al. | A new feature extraction algorithm based on entropy cloud characteristics of communication signals | |
CN112911597B (en) | Internet of things physical layer multilevel feature extraction method based on radio frequency signal fine portrait | |
Yang et al. | Radio frequency fingerprint recognition method based on generative adversarial net | |
CN113343874A (en) | Large-scale radio signal identification method based on deep convolutional neural network | |
Ying et al. | Differential complex-valued convolutional neural network-based individual recognition of communication radiation sources | |
Zeng et al. | Multi-channel attentive feature fusion for radio frequency fingerprinting | |
Haji Bagheri Fard et al. | Rogue device discrimination in ZigBee networks using wavelet transform and autoencoders | |
Hao et al. | Contrastive self-supervised clustering for specific emitter identification | |
Germain et al. | Channel prediction and transmitter authentication with adversarially-trained recurrent neural networks | |
Xu et al. | A jamming recognition algorithm based on deep neural network in satellite navigation system | |
Zhang et al. | Bluetooth Low Energy Device Identification Based on Link Layer Broadcast Packet Fingerprinting | |
Zhang et al. | Wi-Fi device identification based on multi-domain physical layer fingerprint | |
Li et al. | Adaptive filter-based approach for phy fingerprints extraction of fast ethernet network | |
CN112633121A (en) | Radiation source identification method based on Hilbert transform and multi-fractal dimension characteristics | |
CN113065385A (en) | Communication radiation source individual identification method based on axial integral bispectrum | |
Zhang et al. | Triplet Network and Unsupervised Clustering Based Zero-Shot Radio Frequency Fingerprint Identification With Extremely Small Sample Size |
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 |