CN117768884A - Zero-power-consumption communication authentication method, device, equipment and medium based on radio frequency fingerprint - Google Patents
Zero-power-consumption communication authentication method, device, equipment and medium based on radio frequency fingerprint Download PDFInfo
- Publication number
- CN117768884A CN117768884A CN202311658621.4A CN202311658621A CN117768884A CN 117768884 A CN117768884 A CN 117768884A CN 202311658621 A CN202311658621 A CN 202311658621A CN 117768884 A CN117768884 A CN 117768884A
- Authority
- CN
- China
- Prior art keywords
- radio frequency
- zero
- data
- terminal
- power consumption
- 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.)
- Pending
Links
- 230000006854 communication Effects 0.000 title claims abstract description 51
- 238000004891 communication Methods 0.000 title claims abstract description 50
- 238000000034 method Methods 0.000 title claims abstract description 47
- 238000012549 training Methods 0.000 claims abstract description 95
- 238000012360 testing method Methods 0.000 claims abstract description 60
- 239000013598 vector Substances 0.000 claims abstract description 29
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 26
- 238000001514 detection method Methods 0.000 claims description 41
- 238000012545 processing Methods 0.000 claims description 22
- 238000004590 computer program Methods 0.000 claims description 18
- QVFWZNCVPCJQOP-UHFFFAOYSA-N chloralodol Chemical compound CC(O)(C)CC(C)OC(O)C(Cl)(Cl)Cl QVFWZNCVPCJQOP-UHFFFAOYSA-N 0.000 claims description 13
- 230000006870 function Effects 0.000 claims description 13
- 238000003860 storage Methods 0.000 claims description 11
- 238000011176 pooling Methods 0.000 claims description 10
- 238000000605 extraction Methods 0.000 claims description 6
- 230000000007 visual effect Effects 0.000 claims description 6
- ORILYTVJVMAKLC-UHFFFAOYSA-N Adamantane Natural products C1C(C2)CC3CC1CC2C3 ORILYTVJVMAKLC-UHFFFAOYSA-N 0.000 claims description 5
- 230000004913 activation Effects 0.000 claims description 5
- 230000000087 stabilizing effect Effects 0.000 claims description 5
- 238000013507 mapping Methods 0.000 claims description 4
- 238000003306 harvesting Methods 0.000 claims description 2
- 230000007246 mechanism Effects 0.000 abstract description 11
- 238000010586 diagram Methods 0.000 description 12
- 238000005516 engineering process Methods 0.000 description 7
- 230000008569 process Effects 0.000 description 5
- 238000010276 construction Methods 0.000 description 4
- 230000005540 biological transmission Effects 0.000 description 2
- 238000005562 fading Methods 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000000737 periodic effect Effects 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- DWDGSKGGUZPXMQ-UHFFFAOYSA-N OPPO Chemical compound OPPO DWDGSKGGUZPXMQ-UHFFFAOYSA-N 0.000 description 1
- 238000010521 absorption reaction Methods 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000009977 dual effect Effects 0.000 description 1
- 230000008014 freezing Effects 0.000 description 1
- 238000007710 freezing Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000010200 validation analysis Methods 0.000 description 1
Classifications
-
- 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
Abstract
The invention belongs to the technical field of network communication, and particularly relates to a zero-power-consumption communication authentication method, device, equipment and medium based on radio frequency fingerprints, which comprise the steps of collecting original radio frequency data from a zero-power-consumption terminal, and dividing a training set and a testing set; constructing an EfficientNet depth feature extractor based on initial terminal features corresponding to the original radio frequency data; extracting depth feature vectors of the training set, defining an optimal threshold for identifying malicious equipment, and training by combining a KNN algorithm to obtain a lightweight authentication model; and inputting the test set into a trained lightweight authentication model, and outputting an authentication result of the source of the data packet. The zero-power consumption communication security scheme provided by the invention has the advantages of strong expansibility and small calculated amount of an identity authentication mechanism, and is suitable for solving the authentication problem of a zero-power consumption communication terminal.
Description
Technical Field
The invention belongs to the technical field of network communication, and particularly relates to a zero-power consumption communication authentication method, device, equipment and medium based on radio frequency fingerprints.
Background
On day 1 and 20 of 2022, the OPPO institute issues "white paper for zero power communications", which uses radio frequency energy harvesting, back scattering, and low power operation techniques. The zero-power consumption communication terminal is provided with a tiny energy collector and can collect radio frequency energy in the air, so that the zero-power consumption communication terminal has the characteristic of no battery. In addition, in the zero-power consumption system, the transmitting terminal encodes information by using an extremely simple circuit structure and communicates by using a back scattering technology, so that the communication cost is greatly reduced, and the communication requirement of the Internet of things in the future can be met. In view of the excellent characteristics, the zero-power consumption communication is expected to develop into the technology of the next generation of Internet of things.
As with other internet of things communication scenarios, trusted access and secure data transmission remain important in a zero power consumption communication scenario. However, due to the extremely limited software and hardware capabilities, extremely small memory capacity, extremely low power consumption budget, and the dependence of the communication process on external energy supply of the zero-power terminal, conventional security mechanisms, such as encryption mechanisms, digital signatures, and the like, are not suitable for trusted access and secure transmission in the zero-power scenario. Therefore, there is a need to study lightweight security mechanisms suitable for zero power consumption communications.
Radio frequency fingerprints are used as an additional security layer for wireless devices that use unique fingerprints to identify wireless devices for protection against fraud and simulated attacks. The fingerprint information of the device comes from hardware defects in the manufacturing process, and even the same product produced by the same product line has differences in radio frequency fingerprints. Radio frequency fingerprinting is used in many fields, such as intrusion detection, quantum computation, channel estimation, etc.
Based on the radio frequency fingerprint technology, different zero-power consumption communication nodes can be distinguished by utilizing the slight difference of the radio frequency signals of the transmitters of the zero-power consumption communication nodes, the radio frequency fingerprint information of the node equipment is extracted, the processing, the calculation and the identification are performed, the identity information of the zero-power consumption communication nodes is verified, and malicious equipment and legal equipment are identified by utilizing the physical information of the terminal. Compared with the traditional security mechanism with an encryption algorithm as a core, the authentication mode of the physical layer is realized by the radio frequency fingerprint technology without encrypting and decrypting the information, and the authentication mode is interactively transferred from the terminal to the central processing unit for unified authentication, so that the authentication mode has the characteristics of small calculated amount, low power consumption and the like, and is suitable for researching the security mechanism of zero-power-consumption communication.
Disclosure of Invention
the application aims to provide a zero-power consumption communication authentication method, device, equipment and medium based on radio frequency fingerprints, which are used for solving the problem that the traditional authentication mechanism is not applicable due to the limited software and hardware capabilities of the zero-power consumption communication equipment in the prior art.
In order to achieve the above purpose, the present invention adopts the following technical scheme:
In a first aspect, the present invention provides a zero power consumption communication authentication method based on radio frequency fingerprint, including:
Collecting radio frequency data of a zero power consumption terminal to be authenticated;
inputting the collected radio frequency data of the zero-power consumption terminal to be authenticated into a pre-trained lightweight authentication model, and outputting an authentication result of a data packet source;
The pre-trained lightweight authentication model is obtained through training by the following steps: collecting original radio frequency data from a zero-power-consumption terminal, and dividing a training set and a testing set; constructing an EfficientNet depth feature extractor based on initial terminal features corresponding to the original radio frequency data; extracting depth feature vectors of the training set, defining an optimal threshold for identifying malicious equipment, and training by combining a KNN algorithm to obtain a lightweight authentication model; and inputting the test set into a lightweight authentication model obtained by training, and obtaining a lightweight authentication model trained in advance by the test.
Specifically, the acquiring the original radio frequency data from the zero power consumption terminal, and dividing the training set and the testing set includes:
the method comprises the steps of combining a carrier transmitter and a signal receiver, and collecting original radio frequency data of a zero-power-consumption terminal;
storing a data packet containing original radio frequency data from a zero power consumption terminal;
Dividing the data packet into a training set and a testing set based on a preset dividing proportion;
the data in the training set are data which are collected in a point-to-point, fixed-length, visual and equidistant mode; the test set data sources are indoor and outdoor, including data in different distances, visible and invisible environments.
Specifically, the zero-power consumption terminal comprises a radio frequency switch, a radio frequency energy collection module, a power management module and a signal processing module;
The radio frequency energy collection module is used for converting a carrier signal sent by the transmitter into a direct current signal, storing energy through the power management module and providing voltage stabilizing output for the signal processing module; when the terminal is not activated, the radio frequency switch is in an off state, and when the terminal is activated, the signal processing module encodes the acquired signal according to the LoRa protocol and controls the state of the radio frequency switch to modulate the carrier signal;
The signal receiver demodulates and decodes the signal according to the LoRa protocol, and restores the signal acquired by the zero-power consumption terminal.
specifically, the acquiring the original radio frequency data from the zero power consumption terminal, after dividing the training set and the testing set, further includes: and intercepting target fields of the training set and the testing set, intercepting signals with the same length of the Raw-I/Q signals, and generating initial terminal characteristics corresponding to the original radio frequency data by combining the non-zero parts of the zero-power consumption terminal characteristics as the initial characteristics.
Specifically, the construction of the Efficient Net depth feature extractor comprises the following steps:
Selecting an EfficientNet-B0 network structure, adding a global pooling layer at the tail end of the EfficientNet-B0, and converting the space dimension of a feature map formed by initial terminal features into a vector with a fixed length;
Adding a full connection layer after the global pooling layer, and mapping the feature vectors contained in the feature map to a 512-dimensional output space;
selecting a ReLU as an activation function, introducing nonlinearity, applying L2 regularization to control the complexity of a model, and training a depth feature extractor by using an Adam optimizer;
a vector of 512 elements is output.
Specifically, the lightweight authentication model is used for malicious equipment detection and equipment classification; the malicious device detection determines whether the transmitter belongs to a legal group, and if not, the tag of the device is further inferred by device classification.
Specifically, the performing depth feature vector extraction on the training set, and defining the optimal threshold for identifying the malicious device includes:
selecting training set data of a plurality of legal devices, and obtaining K legal depth features with labels through an effective Net feature extractor;
Extracting depth features of the rest training set data, and calculating average distances between the depth features and N adjacent features in the RFF feature database to serve as detection scores;
searching an optimal threshold value by using the residual training set data; when the detection score is higher than a predefined detection threshold beta, the data is considered to be from an illegal device, and vice versa.
Specifically, the average distance to the N nearest neighbor features in the RFF feature database is determined by:
wherein D isiIs the euclidean distance to the ith neighbor.
Specifically, the step of inputting the test set into the trained lightweight authentication model and outputting the authentication result of the data packet source includes: when determining any data packet source in the test set, firstly obtaining depth characteristics through an RFF extractor, and calculating the detection score of the depth characteristics;
Judging illegal equipment according to the detection score; the KNN algorithm taking majority vote, which is higher than the predefined detection threshold β, is considered as an illegal device, and the KNN algorithm taking majority vote, which is lower than the predefined detection threshold β, is attributed to the device where N neighbors appear the most, and if the neighbors are the same, the shortest distance is considered.
In a second aspect, the present invention provides a zero power consumption communication authentication device based on radio frequency fingerprint, including:
the acquisition module is used for acquiring radio frequency data of the zero-power consumption terminal to be authenticated;
The authentication module is used for inputting the acquired radio frequency data of the zero-power consumption terminal to be authenticated into a pre-trained lightweight authentication model and outputting an authentication result of a data packet source; the pre-trained lightweight authentication model is obtained through training by the following steps: collecting original radio frequency data from a zero-power-consumption terminal, and dividing a training set and a testing set; constructing an EfficientNet depth feature extractor based on initial terminal features corresponding to the original radio frequency data; extracting depth feature vectors of the training set, defining an optimal threshold for identifying malicious equipment, and training by combining a KNN algorithm to obtain a lightweight authentication model; and inputting the test set into a lightweight authentication model obtained by training, and obtaining a lightweight authentication model trained in advance by the test.
In a third aspect, the present invention provides an electronic device comprising a processor and a memory, the processor being configured to execute a computer program stored in the memory to implement a zero power consumption communication authentication method based on radio frequency fingerprint as described above.
in a fourth aspect, the present invention provides a computer readable storage medium storing at least one instruction that when executed by a processor implements a zero power consumption communication authentication method based on radio frequency fingerprint as described above.
Compared with the prior art, the invention has the following beneficial effects:
The invention provides a zero-power consumption communication authentication method based on a radio frequency fingerprint technology, which solves the problem that the traditional authentication mechanism is not applicable due to the limited software and hardware capability of zero-power consumption communication equipment.
In the invention, the original Raw-I/Q data of the equipment is collected, and an identity authentication mechanism of the face-phase zero-power-consumption terminal using a lightweight EfficientNet network and a KNN algorithm is constructed, so that the invention is a zero-power-consumption communication security scheme with high use value on a physical layer, and the identity authentication mechanism has strong expansibility and small calculated amount, and is suitable for solving the authentication of the zero-power-consumption communication terminal.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this specification, illustrate embodiments of the application and together with the description serve to explain the application. In the drawings:
FIG. 1 is a flowchart of a zero power consumption communication authentication method based on radio frequency fingerprints provided by an embodiment of the invention;
FIG. 2 is a flowchart of another method for authenticating zero-power communication based on a radio frequency fingerprint according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of a zero power terminal provided by an embodiment of the present invention;
Fig. 4 is a schematic structural diagram of a dual station LoRa system according to an embodiment of the present invention;
FIG. 5 is a schematic view of an acquisition environment provided by an embodiment of the present invention;
FIG. 6 is a block diagram of a depth feature extractor model provided by an embodiment of the present invention;
Fig. 7 is a block diagram of a zero-power consumption communication authentication device based on a radio frequency fingerprint according to an embodiment of the present invention;
FIG. 8 is a block diagram of another apparatus for authenticating zero-power communication based on RF fingerprints according to an embodiment of the present invention;
Fig. 9 is a block diagram of an electronic device according to an embodiment of the present invention.
Detailed Description
embodiments of the present invention will be described in detail below with reference to the accompanying drawings in conjunction with the embodiments.
the application will be described in detail below with reference to the drawings in connection with embodiments. It should be noted that, without conflict, the embodiments of the present application and features of the embodiments may be combined with each other.
The following detailed description is exemplary and is intended to provide further details of the application. Unless defined otherwise, all technical terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of example embodiments in accordance with the application.
Example 1
the embodiment 1 of the invention provides a zero-power consumption communication authentication method based on radio frequency fingerprints, which is shown in fig. 1 and comprises the following steps:
S101, acquiring radio frequency data of a zero-power-consumption terminal to be authenticated;
s102, inputting the collected radio frequency data of the zero-power consumption terminal to be authenticated into a pre-trained lightweight authentication model, and outputting an authentication result of a data packet source;
The pre-trained lightweight authentication model is obtained through training by the following steps: collecting original radio frequency data from a zero-power-consumption terminal, and dividing a training set and a testing set; constructing an EfficientNet depth feature extractor based on initial terminal features corresponding to the original radio frequency data; extracting depth feature vectors of the training set, defining an optimal threshold for identifying malicious equipment, and training by combining a KNN algorithm to obtain a lightweight authentication model; and inputting the test set into a lightweight authentication model obtained by training, and obtaining a lightweight authentication model trained in advance by the test.
in one embodiment: the acquiring the original radio frequency data from the zero-power consumption terminal, and dividing the training set and the testing set comprises:
the method comprises the steps of combining a carrier transmitter and a signal receiver, and collecting original radio frequency data of a zero-power-consumption terminal;
storing a data packet containing original radio frequency data from a zero power consumption terminal;
Dividing the data packet into a training set and a testing set based on a preset dividing proportion;
the data in the training set are data which are collected in a point-to-point, fixed-length, visual and equidistant mode; the test set data sources are indoor and outdoor, including data in different distances, visible and invisible environments.
In one embodiment: the zero-power consumption terminal comprises a radio frequency switch, a radio frequency energy collection module, a power management module and a signal processing module;
The radio frequency energy collection module is used for converting a carrier signal sent by the transmitter into a direct current signal, storing energy through the power management module and providing voltage stabilizing output for the signal processing module; when the terminal is not activated, the radio frequency switch is in an off state, and when the terminal is activated, the signal processing module encodes the acquired signal according to the LoRa protocol and controls the state of the radio frequency switch to modulate the carrier signal;
The signal receiver demodulates and decodes the signal according to the LoRa protocol, and restores the signal acquired by the zero-power consumption terminal.
In one embodiment: the method for acquiring the original radio frequency data from the zero-power-consumption terminal further comprises the following steps of: and intercepting target fields of the training set and the testing set, intercepting signals with the same length of the Raw-I/Q signals, and generating initial terminal characteristics corresponding to the original radio frequency data by combining the non-zero parts of the zero-power consumption terminal characteristics as the initial characteristics.
in one embodiment: the construction of the Efficient Net depth feature extractor comprises the following steps:
Selecting an EfficientNet-B0 network structure, adding a global pooling layer at the tail end of the EfficientNet-B0, and converting the space dimension of a feature map formed by initial terminal features into a vector with a fixed length;
Adding a full connection layer after the global pooling layer, and mapping the feature vectors contained in the feature map to a 512-dimensional output space;
selecting a ReLU as an activation function, introducing nonlinearity, applying L2 regularization to control the complexity of a model, and training a depth feature extractor by using an Adam optimizer;
a vector of 512 elements is output.
in one embodiment: the lightweight authentication model is used for malicious equipment detection and equipment classification; the malicious device detection determines whether the transmitter belongs to a legal group, and if not, the tag of the device is further inferred by device classification.
In one embodiment: the step of extracting depth feature vectors of the training set, and the step of defining an optimal threshold for identifying malicious equipment comprises the following steps:
selecting training set data of a plurality of legal devices, and obtaining K legal depth features with labels through an effective Net feature extractor;
Extracting depth features of the rest training set data, and calculating average distances between the depth features and N adjacent features in the RFF feature database to serve as detection scores;
searching an optimal threshold value by using the residual training set data; when the detection score is higher than a predefined detection threshold beta, the data is considered to be from an illegal device, and vice versa.
in one embodiment: the average distance to the N nearest neighbor features in the RFF feature database is determined by:
wherein D isiIs the euclidean distance to the ith neighbor.
In one embodiment: the step of inputting the collected radio frequency data of the zero-power consumption terminal to be authenticated into a pre-trained lightweight authentication model, and the step of outputting the authentication result of the data packet source comprises the following steps: when determining any data packet source in the radio frequency data, firstly obtaining depth characteristics through an RFF extractor, and calculating the detection score of the depth characteristics;
Judging illegal equipment according to the detection score; the KNN algorithm taking majority vote, which is higher than the predefined detection threshold β, is considered as an illegal device, and the KNN algorithm taking majority vote, which is lower than the predefined detection threshold β, is attributed to the device where N neighbors appear the most, and if the neighbors are the same, the shortest distance is considered.
example 2:
the embodiment 2 of the invention provides a zero-power consumption communication authentication method based on radio frequency fingerprints, which is shown in fig. 2 and comprises the following steps:
S1, acquiring original radio frequency data from a zero-power-consumption terminal, and dividing a training set and a testing set;
S2, constructing an EfficientNet depth feature extractor based on initial terminal features corresponding to the original radio frequency data;
s3, extracting depth feature vectors of the training set, defining an optimal threshold for identifying malicious equipment, and training by combining a KNN algorithm to obtain a lightweight authentication model;
S4, inputting the test set into a trained lightweight authentication model, and outputting an authentication result of a data packet source; the test is performed by obtaining a pre-trained lightweight authentication model.
S5, acquiring radio frequency data of a zero power consumption terminal to be authenticated; and inputting the acquired radio frequency data of the zero-power consumption terminal to be authenticated into a pre-trained lightweight authentication model, and outputting an authentication result of the source of the data packet.
as shown in fig. 3, the zero power consumption terminal in step S1 includes a radio frequency switch, a radio frequency energy collecting module, a power management module and a signal processing module;
The radio frequency energy collection module is used for converting a carrier signal sent by the transmitter into a direct current signal, storing energy through the power management module and providing voltage stabilizing output for the signal processing module; when the terminal is not activated, the radio frequency switch is in an off state, and when the terminal is activated, the signal processing module encodes the acquired signal according to the LoRa protocol and controls the state of the radio frequency switch to modulate the carrier signal;
The signal receiver demodulates and decodes the signal according to the LoRa protocol, and restores the signal acquired by the zero-power consumption terminal.
In step S1, original radio frequency data of a zero-power-consumption terminal, namely original Raw-I/Q data of equipment; before executing step S1, it is necessary to implement a dual-station LoRa-based zero-power consumption communication system, and build a test platform to test connectivity of the system. The built test platform comprises 1 transmitter, 1 receiver and 16 zero-power-consumption terminals, wherein 12 terminal devices are marked as legal devices (marked as 0-11), and 4 terminal devices are marked as illegal devices (marked as 12-15). The specific acquisition mode is as follows: the radio frequency source (transmitter) transmits radio frequency signals, the terminal uses part of the received radio frequency signals for coding data packets (payload data in the data packets come from the sensor), the other part of the received radio frequency signals are used for reflecting modulated waves, the receiver receives back scattering signals of the terminal, and the back scattering signals are connected with the PC end through a singlechip on the receiver, so that the data packets are stored and the signals are decoded.
The zero-power consumption communication system of the double station LoRa is specifically designed:
As shown in fig. 4, the transmitter broadcasts a carrier signal to the air, the zero-power consumption terminal receives the carrier, determines according to the state of the radio frequency switch, when the terminal is not activated, the radio frequency switch is in an off state, the carrier enters the radio frequency energy collection module and is converted into a direct current signal, and the power management module stores energy and provides voltage stabilizing output for the signal processing module. And after the terminal is activated, the signal processing module encodes the acquired signal into a baseband signal according to the LoRa protocol, and controls the state of the radio frequency switch to modulate the carrier signal. When the radio frequency switch is in a conducting state, the terminal reflects a carrier signal sent by the transmitter, and the terminal can send a high-level carrier equivalently; when the radio frequency switch is in an off state, the radio frequency energy collection module in the terminal converts the carrier signal sent by the transmitter into a direct current signal, and the radio frequency energy collection module can be equivalent to the terminal to send out a low-level carrier. The receiver demodulates and decodes the modulated signal according to the LoRa protocol, thereby recovering the signal collected by the terminal.
In step S1, collecting original radio frequency data from a zero power consumption terminal, and dividing a training set and a testing set includes:
the method comprises the steps of combining a carrier transmitter and a signal receiver, and collecting original radio frequency data of a zero-power-consumption terminal;
storing a data packet containing original radio frequency data from a zero power consumption terminal;
Dividing the data packet into a training set and a testing set based on a preset dividing proportion;
the data in the training set are data which are collected in a point-to-point, fixed-length, visual and equidistant mode; the test set data sources are indoor and outdoor, including data in different distances, visible and invisible environments.
Specifically, rx in fig. 5 is the receiver position, and data acquisition is performed for three consecutive days, where the ratio of data packets in the training set to the test set is 5:2. The data in the training set are all data acquisition performed under the conditions of point-to-point, fixed-length and visual equidistance, the data sources of the testing set are indoor and outdoor, the testing set comprises data under different distances, visual environments and invisible environments, in order to generate more training data, the performance of the RFF extractor is improved, meanwhile, the cost of data collection is reduced, in the embodiment, a data enhancement technology based on an ITU-R channel model is introduced, and the technology firstly utilizes the ITU-R model to create an FIR filter. The FIR model is represented by a finite M-sequence filter, i.e., phi = { phi1，φ2，...，φM}. Given inputwhere N is a set of consecutive I/Q samples that serve as inputs to the classifier. For/>filtered nth element/>Can be written as:
φjis a set of weights, expressed as:
Wherein T issis the channel of the input sample, τkis the relative delay of the path which is not less than 1 and not more than K, K is the total number of multipath fading channels, akis a complex path multipath fading channel, calculated using a sine sum.
Preferably, after collecting the original radio frequency data from the zero power consumption terminal and dividing the training set and the testing set, the method further comprises: and intercepting target fields of the training set and the testing set, intercepting signals with the same length of the Raw-I/Q signals, and generating initial terminal characteristics corresponding to the original radio frequency data by combining the non-zero parts of the zero-power consumption terminal characteristics as the initial characteristics.
the specific operation of the target field interception is as follows: the predefined threshold is 1/8 of the OOK signal amplitude, when the M n threshold is reached, the packet is dropped, otherwise. Meanwhile, in order to ensure consistent signal length of the input network, the OOK signal period L is required to be utilizedsPerforming secondary interception, length L of input1A set of non-zero OOK signals between each data packet.
Specifically, the back-scattered signal received by the receiver has only two states, namely the absorption state and the reflection state of the terminal, so that the whole received waveform contains terminal characteristics (including packet intervals), and thus the conventional preamble-based and payload-based characteristic extraction is not suitable for the system of the present invention. By combining the OOK modulation characteristic and the radio frequency fingerprint characteristic without considering the information characteristic, the invention adopts a method for intercepting the periodic effective signal, namely, the interval of the data packet is removed from the 0 part (the device fingerprints contained in the data packet and the data packet are consistent), and the periodic effective waveform is reserved. The Schmidl & Cox algorithm is used in this embodiment to extract the effective wave as input to the feature extractor, expressed as:
Wherein r is*n+k represents the conjugate of r n+k, L is a symbol length, and a threshold is predefined to detect the arrival of the effective wave: when the M [ n ] > threshold is reached, the packet arrives, and otherwise the signal is discarded.
After the target field is intercepted for each device, integrating the intercepted I/Q signals, and in order to ensure the initial characteristic field of the input RFF extractor, the OOK signal period L is needed to be utilizedsPerforming secondary interception, length L of input1The following expression needs to be satisfied:
L1＝n1Ts
in step S3, the construction of the afflicientnet depth feature extractor includes:
As shown in fig. 6, selecting an afflicientnet-B0 network structure, adding a global pooling layer at the end of the afflicientnet-B0, and converting the spatial dimension of a feature map formed by the initial terminal features into a vector with a fixed length;
Adding a full connection layer after the global pooling layer, and mapping the feature vectors contained in the feature map to a 512-dimensional output space;
selecting a ReLU as an activation function, introducing nonlinearity, applying L2 regularization to control the complexity of a model, and training a depth feature extractor by using an Adam optimizer;
a vector of 512 elements is output.
Specifically, since the original I/Q signal includes two rows, one row is in-phase acquisition and the other row is quadrature acquisition, after each data packet is subjected to target field interception, both signals are input into the network in the form of data. Thus, in building the EfficientNet-B0 feature extractor, the pre-trained EfficientNet-B0 model is first loaded in this embodiment, and include_top=false is set to remove the top classifier of the model (training feature extractor, reducing the number of parameters). Then a global averaging pooling layer is added at the end of Efficient Net-B0 to convert the spatial dimension of the feature map into a fixed length vector. A fully connected layer is added after the global pooling layer to map feature vectors to 512-dimensional output space. The ReLU is selected as the activation function and then the weights of EfficientNet-B0 are frozen (the pre-trained EfficientNet-B0 model has learned generic feature representations over large scale data that can be used for various image classification tasks by freezing the weights, the feature extraction capabilities of the pre-trained model can be preserved while only the custom full-connection layer is trained to accommodate a particular task) so that it does not participate in the training. Then setting a learning rate and an optimizer, and compiling a model: the Adam optimizer was used and the learning rate was set to 0.001 and the loss function was set to the class cross entropy. Next, a termination condition is set, here using Early Stopping strategy, and training will be terminated Early and the best weights restored when the loss of validation sets is not improved in the consecutive 3 epochs. Finally, a trained feature extractor is used for extracting depth feature vectors of the data packets.
In step S3, performing depth feature vector extraction on the training set, and defining an optimal threshold for identifying malicious devices includes:
selecting training set data of a plurality of legal devices, and obtaining K legal depth features with labels through an effective Net feature extractor;
Extracting depth features of the rest training set data, and calculating average distances between the depth features and N adjacent features in the RFF feature database to serve as detection scores;
searching an optimal threshold value by using the residual training set data; when the detection score is higher than a predefined detection threshold beta, the data is considered to be from an illegal device, and vice versa.
Further, the ROC curve is utilized to find an optimal threshold β for malicious device identification. The feature extractor training uses partially labeled data (k=12) in the training set, and the remaining training set data is used to find the optimal malicious threshold. The step of finding the optimal threshold β is: sequentially passing the rest data (with labels and 16 classes) in the training set through an RFF extractor to obtain depth characteristics of each data, and calculating detection score D of each dataavg
Wherein D isiis the euclidean distance to the ith neighbor. The method is characterized in that an ROC curve is utilized to find the optimal threshold value for malicious equipment identification, the process is a two-classification process, and legal equipment classification is a multi-classification process. And marking the tag devices No. 0-11 as legal classes and No. 12-15 as malicious classes. Sequentially sending the data in the test set to an RFF extractor, sequentially passing the rest data in the training set through the RFF extractor to obtain depth characteristics of each data, and calculating the detection score D of each dataavg. The ROC curve is utilized to find the optimal threshold, and samples in the confusion evidence are divided into four categories of True examples TP (True Position), false positive examples FP (False Position), false Negative examples FN (False Negative) and True Negative examples (TN), and the definition of the four categories is: the true examples represent the legal class number correctly judged as legal class by the algorithm; the false positive example represents the number of malicious classes judged as legal by the algorithm; false counter examples represent the legal class number misjudged as malicious class by the algorithm; true and false examples represent the number of malicious classes that are correctly judged by the algorithm to be malicious classes. And calculating TPR and FPR under different thresholds according to the prediction result and the real labels of the test set, and determining an optimal value classification threshold by maximizing the TPR and minimizing the FPR.
the lightweight authentication model in the step S3 is built based on KNN algorithm training and is used for malicious equipment detection and equipment classification; the malicious device detection determines whether the transmitter belongs to a legal group, and if not, the tag of the device is further inferred by device classification.
Finally, step S4 inputs the test set into the trained lightweight authentication model, and the authentication result of the output data packet source comprises: when determining any data packet source in the test set, firstly obtaining depth characteristics through an RFF extractor, and calculating the detection score of the depth characteristics;
Judging illegal equipment according to the detection score; the KNN algorithm taking majority vote, which is higher than the predefined detection threshold β, is considered as an illegal device, and the KNN algorithm taking majority vote, which is lower than the predefined detection threshold β, is attributed to the device where N neighbors appear the most, and if the neighbors are the same, the shortest distance is considered.
Specifically, the test set data identification process is: inputting the data (including the data enhancement part) into an RFF extractor to obtain corresponding depth characteristics, calculating detection scores corresponding to the characteristics, and if the detection scores are higher than a threshold value, considering the data as a data packet corresponding to malicious equipment, and otherwise, classifying legal equipment. In order to determine the attribution of the legal category, the method adopts a majority voting method, namely the data packet is considered to belong to the legal equipment category with the largest occurrence among the nearest N neighbors, and if the data packet is the same, the shortest distance is prioritized. Thus, an identity authentication mechanism of the zero-power consumption communication terminal is established.
Example 3
As shown in fig. 7, based on the same inventive concept as the above embodiment, the present invention further provides a zero power consumption communication authentication device based on radio frequency fingerprint, including:
The acquisition module is used for acquiring the original radio frequency data from the zero-power-consumption terminal and dividing a training set and a testing set;
The construction module is used for constructing an EfficientNet depth feature extractor based on the initial terminal features corresponding to the original radio frequency data;
the extraction module is used for extracting depth feature vectors of the training set, defining an optimal threshold for identifying malicious equipment, and combining KNN algorithm training to obtain a lightweight authentication model;
and the authentication module is used for inputting the test set into the trained lightweight authentication model and outputting the authentication result of the data packet source.
Example 4
as shown in fig. 8, based on the same inventive concept as embodiment 1, the present invention further provides a zero power consumption communication authentication device based on radio frequency fingerprint, comprising:
the acquisition module is used for acquiring radio frequency data of the zero-power consumption terminal to be authenticated;
The authentication module is used for inputting the acquired radio frequency data of the zero-power consumption terminal to be authenticated into a pre-trained lightweight authentication model and outputting an authentication result of a data packet source; the pre-trained lightweight authentication model is obtained through training by the following steps: collecting original radio frequency data from a zero-power-consumption terminal, and dividing a training set and a testing set; constructing an EfficientNet depth feature extractor based on initial terminal features corresponding to the original radio frequency data; extracting depth feature vectors of the training set, defining an optimal threshold for identifying malicious equipment, and training by combining a KNN algorithm to obtain a lightweight authentication model; and inputting the test set into a lightweight authentication model obtained by training, and obtaining a lightweight authentication model trained in advance by the test.
Example 5
As shown in fig. 9, the present invention further provides an electronic device 100 for implementing the radio frequency fingerprint-based zero power consumption communication authentication method according to the embodiment; the electronic device 100 comprises a memory 101, at least one processor 102, a computer program 103 stored in the memory 101 and executable on the at least one processor 102, and at least one communication bus 104.
The memory 101 may be used to store a computer program 103, and the processor 102 implements the steps of a zero power consumption communication authentication method based on radio frequency fingerprint of embodiment 1 by running or executing the computer program stored in the memory 101 and invoking data stored in the memory 101.
The memory 101 may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program (such as a sound playing function, an image playing function, etc.) required for at least one function, and the like; the storage data area may store data (such as audio data) created according to the use of the electronic device 100, and the like. In addition, the memory 101 may include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash Card (Flash Card), at least one disk storage device, a Flash memory device, or other non-volatile solid state storage device.
The at least one processor 102 may be a central processing unit (Central Processing Unit, CPU), but may also be other general purpose processors, digital signal processors (Digital Signal Processor, DSP), application specific integrated circuits (Application Specific Integrated Circuit, ASIC), field-programmable gate arrays (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or the like. The processor 102 may be a microprocessor or the processor 102 may be any conventional processor or the like, the processor 102 being a control center of the electronic device 100, the various interfaces and lines being utilized to connect various portions of the overall electronic device 100.
The memory 101 in the electronic device 100 stores a plurality of instructions to implement a zero power consumption communication authentication method based on radio frequency fingerprint, and the processor 102 may execute the plurality of instructions to implement:
Collecting radio frequency data of a zero power consumption terminal to be authenticated;
inputting the collected radio frequency data of the zero-power consumption terminal to be authenticated into a pre-trained lightweight authentication model, and outputting an authentication result of a data packet source;
The pre-trained lightweight authentication model is obtained through training by the following steps: collecting original radio frequency data from a zero-power-consumption terminal, and dividing a training set and a testing set; constructing an EfficientNet depth feature extractor based on initial terminal features corresponding to the original radio frequency data; extracting depth feature vectors of the training set, defining an optimal threshold for identifying malicious equipment, and training by combining a KNN algorithm to obtain a lightweight authentication model; and inputting the test set into a lightweight authentication model obtained by training, and obtaining a lightweight authentication model trained in advance by the test.
Example 6
The modules/units integrated with the electronic device 100 may be stored in a computer readable storage medium if implemented in the form of software functional units and sold or used as a stand alone product. Based on such understanding, the present invention may implement all or part of the flow of the method of the above embodiment, or may be implemented by a computer program to instruct related hardware, where the computer program may be stored in a computer readable storage medium, and when the computer program is executed by a processor, the steps of each method embodiment described above may be implemented. Wherein the computer program comprises computer program code, which may be in the form of source code, object code, executable files or in some intermediate form, etc. The computer readable medium may include: any entity or device capable of carrying computer program code, a recording medium, a U disk, a removable hard disk, a magnetic disk, an optical disk, a computer Memory, and a Read-Only Memory (ROM).
It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
the present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
in the description of the present specification, the descriptions of the terms "one embodiment," "example," "specific example," and the like, mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the present invention. In this specification, schematic representations of the above terms do not necessarily refer to the same embodiments or examples. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
finally, it should be noted that: the above embodiments are only for illustrating the technical aspects of the present invention and not for limiting the same, and although the present invention has been described in detail with reference to the above embodiments, it should be understood by those of ordinary skill in the art that: modifications and equivalents may be made to the specific embodiments of the invention without departing from the spirit and scope of the invention, which is intended to be covered by the claims.
Claims (12)
1. A zero-power consumption communication authentication method based on radio frequency fingerprints is characterized by comprising the following steps:
Collecting radio frequency data of a zero power consumption terminal to be authenticated;
inputting the collected radio frequency data of the zero-power consumption terminal to be authenticated into a pre-trained lightweight authentication model, and outputting an authentication result of a data packet source;
The pre-trained lightweight authentication model is obtained through training by the following steps: collecting original radio frequency data from a zero-power-consumption terminal, and dividing a training set and a testing set; constructing an EfficientNet depth feature extractor based on initial terminal features corresponding to the original radio frequency data; extracting depth feature vectors of the training set, defining an optimal threshold for identifying malicious equipment, and training by combining a KNN algorithm to obtain a lightweight authentication model; and inputting the test set into a lightweight authentication model obtained by training, and obtaining a lightweight authentication model trained in advance by the test.
2. the method of claim 1, wherein the acquiring raw radio frequency data from the zero power terminal, dividing the training set and the test set comprises:
the method comprises the steps of combining a carrier transmitter and a signal receiver, and collecting original radio frequency data of a zero-power-consumption terminal;
storing a data packet containing original radio frequency data from a zero power consumption terminal;
Dividing the data packet into a training set and a testing set based on a preset dividing proportion;
the data in the training set are data which are collected in a point-to-point, fixed-length, visual and equidistant mode; the test set data sources are indoor and outdoor, including data in different distances, visible and invisible environments.
3. the method of claim 2, wherein the zero power terminal comprises a radio frequency switch, a radio frequency energy harvesting module, a power management module, and a signal processing module;
The radio frequency energy collection module is used for converting a carrier signal sent by the transmitter into a direct current signal, storing energy through the power management module and providing voltage stabilizing output for the signal processing module; when the terminal is not activated, the radio frequency switch is in an off state, and when the terminal is activated, the signal processing module encodes the acquired signal according to the LoRa protocol and controls the state of the radio frequency switch to modulate the carrier signal;
The signal receiver demodulates and decodes the signal according to the LoRa protocol, and restores the signal acquired by the zero-power consumption terminal.
4. The method of claim 1, wherein the collecting the raw rf data from the zero-power terminal, after dividing the training set and the test set, further comprises: and intercepting target fields of the training set and the testing set, intercepting signals with the same length of the Raw-I/Q signals, and generating initial terminal characteristics corresponding to the original radio frequency data by combining the non-zero parts of the zero-power consumption terminal characteristics as the initial characteristics.
5. The method of claim 1, wherein the constructing of the EfficientNet depth feature extractor comprises:
Selecting an EfficientNet-B0 network structure, adding a global pooling layer at the tail end of the EfficientNet-B0, and converting the space dimension of a feature map formed by initial terminal features into a vector with a fixed length;
Adding a full connection layer after the global pooling layer, and mapping the feature vectors contained in the feature map to a 512-dimensional output space;
selecting a ReLU as an activation function, introducing nonlinearity, applying L2 regularization to control the complexity of a model, and training a depth feature extractor by using an Adam optimizer;
a vector of 512 elements is output.
6. The method of claim 1, wherein the lightweight authentication model is used for malicious device detection and device classification; the malicious device detection determines whether the transmitter belongs to a legal group, and if not, the tag of the device is further inferred by device classification.
7. The method of claim 1, wherein the depth feature vector extraction of the training set, defining an optimal threshold for identifying malicious devices, comprises:
selecting training set data of a plurality of legal devices, and obtaining K legal depth features with labels through an effective Net feature extractor;
Extracting depth features of the rest training set data, and calculating average distances between the depth features and N adjacent features in the RFF feature database to serve as detection scores;
searching an optimal threshold value by using the residual training set data; when the detection score is higher than a predefined detection threshold beta, the data is considered to be from an illegal device, and vice versa.
8. The method of claim 7, wherein the average distance to N nearest neighbor features in the RFF feature database is determined by:
wherein D isiIs the euclidean distance to the ith neighbor.
9. The method of claim 1, wherein inputting the collected radio frequency data of the to-be-authenticated zero-power consumption terminal into a pre-trained lightweight authentication model, and outputting the authentication result of the data packet source comprises: when determining any data packet source in the radio frequency data, firstly obtaining depth characteristics through an RFF extractor, and calculating the detection score of the depth characteristics;
Judging illegal equipment according to the detection score; the KNN algorithm taking majority vote, which is higher than the predefined detection threshold β, is considered as an illegal device, and the KNN algorithm taking majority vote, which is lower than the predefined detection threshold β, is attributed to the device where N neighbors appear the most, and if the neighbors are the same, the shortest distance is considered.
10. A zero power consumption communication authentication device based on radio frequency fingerprint, comprising:
the acquisition module is used for acquiring radio frequency data of the zero-power consumption terminal to be authenticated;
The authentication module is used for inputting the acquired radio frequency data of the zero-power consumption terminal to be authenticated into a pre-trained lightweight authentication model and outputting an authentication result of a data packet source; the pre-trained lightweight authentication model is obtained through training by the following steps: collecting original radio frequency data from a zero-power-consumption terminal, and dividing a training set and a testing set; constructing an EfficientNet depth feature extractor based on initial terminal features corresponding to the original radio frequency data; extracting depth feature vectors of the training set, defining an optimal threshold for identifying malicious equipment, and training by combining a KNN algorithm to obtain a lightweight authentication model; and inputting the test set into a lightweight authentication model obtained by training, and obtaining a lightweight authentication model trained in advance by the test.
11. An electronic device comprising a processor and a memory, the processor configured to execute a computer program stored in the memory to implement the radio frequency fingerprint based zero power consumption communication authentication method according to any one of claims 1 to 9.
12. A computer readable storage medium storing at least one instruction which when executed by a processor implements the radio frequency fingerprint based zero power consumption communication authentication method according to any one of claims 1 to 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311658621.4A CN117768884A (en) | 2023-12-05 | 2023-12-05 | Zero-power-consumption communication authentication method, device, equipment and medium based on radio frequency fingerprint |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311658621.4A CN117768884A (en) | 2023-12-05 | 2023-12-05 | Zero-power-consumption communication authentication method, device, equipment and medium based on radio frequency fingerprint |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117768884A true CN117768884A (en) | 2024-03-26 |
Family
ID=90324613
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311658621.4A Pending CN117768884A (en) | 2023-12-05 | 2023-12-05 | Zero-power-consumption communication authentication method, device, equipment and medium based on radio frequency fingerprint |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117768884A (en) |
-
2023
- 2023-12-05 CN CN202311658621.4A patent/CN117768884A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Jagannath et al. | A comprehensive survey on radio frequency (rf) fingerprinting: Traditional approaches, deep learning, and open challenges | |
| Zhao et al. | Classification of small UAVs based on auxiliary classifier Wasserstein GANs | |
| Agadakos et al. | Chameleons' oblivion: Complex-valued deep neural networks for protocol-agnostic RF device fingerprinting | |
| Yavuz | Deep learning in cyber security for internet of things | |
| KR102067324B1 (en) | Apparatus and method for analyzing feature of impersonation attack using deep running in wireless wi-fi network | |
| CN112671739B (en) | Node property identification method of distributed system | |
| Li et al. | Channel-based sybil detection in industrial wireless sensor networks: A multi-kernel approach | |
| Agadakos et al. | Deep complex networks for protocol-agnostic radio frequency device fingerprinting in the wild | |
| CN114727286B (en) | Threshold-free multi-attribute physical layer authentication method and related equipment | |
| CN114867018A (en) | Device identification authentication method, system and storage medium using carrier frequency offset | |
| Zeng et al. | Multi-channel attentive feature fusion for radio frequency fingerprinting | |
| CN116776227A (en) | Satellite identity recognition method and device based on feature fusion | |
| CN117768884A (en) | Zero-power-consumption communication authentication method, device, equipment and medium based on radio frequency fingerprint | |
| CN113420791B (en) | Access control method and device for edge network equipment and terminal equipment | |
| CN112566117B (en) | Vehicle node identity recognition method and device based on metric learning | |
| CN115809426A (en) | Radiation source individual identification method and system | |
| Zhang et al. | Bluetooth Low Energy Device Identification Based on Link Layer Broadcast Packet Fingerprinting | |
| CN113765891B (en) | Equipment fingerprint identification method and device | |
| Yin et al. | Detecting CAN overlapped voltage attacks with an improved voltage-based in-vehicle intrusion detection system | |
| Kang et al. | Detecting identity-spoof attack based on BP network in cognitive radio network | |
| WO2021070248A1 (en) | Transmitting device verification device, transmitting device verification system, transmitting device verification method, and computer-readable medium | |
| Alami et al. | Design, analysis and optimization of a microstrip patch antenna at frequency 3.55 GHZ for WiMAX application | |
| CN110912906A (en) | Edge calculation malicious node identification method | |
| Abbas et al. | Radio frequency fingerprinting techniques for device identification: a survey | |
| CN108810030B (en) | Node property identification method based on distributed system |
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 |