CN114978582B - Radio frequency fingerprint identification method and system based on iterative cosine spectrum transformation - Google Patents
Radio frequency fingerprint identification method and system based on iterative cosine spectrum transformation Download PDFInfo
- Publication number
- CN114978582B CN114978582B CN202210376325.4A CN202210376325A CN114978582B CN 114978582 B CN114978582 B CN 114978582B CN 202210376325 A CN202210376325 A CN 202210376325A CN 114978582 B CN114978582 B CN 114978582B
- Authority
- CN
- China
- Prior art keywords
- radio frequency
- transient
- interval
- frequency fingerprint
- cost function
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L63/00—Network architectures or network communication protocols for network security
- H04L63/08—Network architectures or network communication protocols for network security for authentication of entities
- H04L63/0876—Network architectures or network communication protocols for network security for authentication of entities based on the identity of the terminal or configuration, e.g. MAC address, hardware or software configuration or device fingerprint
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/14—Fourier, Walsh or analogous domain transformations, e.g. Laplace, Hilbert, Karhunen-Loeve, transforms
- G06F17/141—Discrete Fourier transforms
- G06F17/142—Fast Fourier transforms, e.g. using a Cooley-Tukey type algorithm
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L27/00—Modulated-carrier systems
- H04L27/0012—Modulated-carrier systems arrangements for identifying the type of modulation
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W12/00—Security arrangements; Authentication; Protecting privacy or anonymity
- H04W12/40—Security arrangements using identity modules
- H04W12/47—Security arrangements using identity modules using near field communication [NFC] or radio frequency identification [RFID] modules
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W12/00—Security arrangements; Authentication; Protecting privacy or anonymity
- H04W12/60—Context-dependent security
- H04W12/69—Identity-dependent
- H04W12/79—Radio fingerprint
-
- 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 provides a radio frequency fingerprint identification method and a system based on iterative cosine spectrum transformation, wherein the method comprises the following steps: the method comprises the steps of performing power envelope and carrier separation on a transmitter signal to be identified; performing first-order derivation on the power envelope, and acquiring a transient signal in a primary transient interval; establishing a floating interval at a starting position and a stopping position of the transient signal based on the continuous length of the transient signal, establishing a fixed step length to set a sliding window to slide in the floating interval, and adjusting the sliding step length until finding the position which minimizes the cost function value when the calculated cost function value is smaller than the cost function value of the transient signal, so as to obtain an optimal target interval; after the signals in the optimal target interval are subjected to fast Fourier transform and discrete cosine transform in sequence, a radio frequency fingerprint feature vector is constructed; inputting the radio frequency fingerprint feature vector into a support vector machine for identity authentication of equipment. The authentication method can accurately authenticate the identity of different infinite devices by utilizing the hardware fine difference between the infinite devices.
Description
Technical Field
The invention relates to the technical field of wireless communication, in particular to a radio frequency fingerprint identification method and system based on iterative cosine spectrum transformation.
Background
With the popularization and development of mobile communication technology, wireless interconnection has been advanced to the aspects of our daily life, and especially with the popularization of the internet of things, great changes will be brought to our life and even production modes. The wireless information interaction mode of electromagnetic waves is utilized, so that the difficulty of networking of mass equipment can be simplified, but the openness of the wireless channel also provides natural conditions for information interception and counterfeiting attack, and becomes the root of the wireless communication security problem. Aiming at the problem of equipment access authentication in a wireless network, password authentication, digital certificate authentication, access authentication based on a blockchain and the like are generally adopted. Essentially, these authentication methods are all based on digital encryption of modern encryption algorithms, and the reliability of access authentication is ensured by utilizing the complexity of the encryption algorithm and the privacy of the initially distributed password. With the continuous improvement of the computing capability of computers in various new systems, the violent decomposition of encryption algorithms becomes easier; in addition, the distribution mechanism of the initial key also has hidden danger of privacy disclosure. Since tolerance effects of electronic components in wireless devices are a main cause of generating radio frequency fingerprints, the radio frequency fingerprints are difficult to clone and have uniqueness, so that the radio frequency fingerprints become a new paradigm of identity authentication of wireless devices.
The existing radio frequency fingerprint technology is mainly steady-state radio frequency fingerprint technology. The steady state signal refers to a signal of the transmitter in a steady power state, mainly from the aspects of modulation mode, noise output difference, stray characteristic difference, and the like of equipment, performs time-frequency analysis on a received signal, extracts physical quantities with independent and steady characteristics as radio frequency fingerprints, and classifies the physical quantities to achieve the purpose of identifying a radiation source. The radio frequency fingerprint method based on the steady-state signal needs more priori knowledge based on the modulation mode, the code pattern design and the like of the target signal, so that the universality is poor.
Disclosure of Invention
According to the technical problem that the prior radio frequency fingerprint needs to be based on prior knowledge such as a modulation mode and a code pattern design of a target signal to cause poor universality, the radio frequency fingerprint identification method and system based on iterative cosine spectrum transformation are provided. The method comprises the steps of separating a power envelope from a carrier wave of a received signal, deeply analyzing the characteristics of the power envelope, converting the power envelope into a space with higher dimensionality by using fast Fourier transform and discrete cosine transform, extracting the energy statistical characteristics of the power envelope to construct a characteristic vector, serving as the radio frequency fingerprint characteristics of equipment, and classifying, namely identifying the equipment according to the characteristics. The invention has wider application range and smaller calculation complexity, and can realize the identification and authentication of the transmitter equipment in practical application.
The invention adopts the following technical means:
a radio frequency fingerprint identification method based on iterative cosine spectrum transformation comprises the following steps:
s1, acquiring a transmitter signal to be identified, and separating a power envelope and a carrier from the transmitter signal to be identified;
s2, after first-order derivation is carried out on the power envelope, a primary transient interval is obtained according to a preset threshold value, and a transient signal is obtained in the primary transient interval;
s3, establishing a floating interval at a start position and a stop position of the transient signal based on the duration of the transient signal, establishing a fixed step length to set a sliding window to slide in the floating interval, continuously calculating the value of a cost function in the sliding process, and adjusting the sliding step length until a position which enables the cost function value to be minimum is found when the calculated cost function value is smaller than the cost function value of the transient signal, so that an optimal target interval is obtained;
s4, after the signal in the optimal target interval is subjected to fast Fourier transform and discrete cosine transform in sequence, a radio frequency fingerprint feature vector is constructed based on a transformation result;
s5, inputting the radio frequency fingerprint feature vector to a support vector machine for identity authentication of equipment.
Further, after the transmitter signal to be identified is acquired in S1, filtering processing is further performed on the transmitter signal to be identified, so as to filter spurious components that are unintentionally acquired in the acquisition process.
Further, establishing a floating interval at a start and stop position of the transient signal based on a duration of the transient signal in S3 includes:
a floating range with the size of w/8 is set at the starting and stopping positions of the primary transient interval, so that a floating interval is constructed, wherein w represents the duration of the transient signal.
Further, the cost function described in S3 is obtained according to the following calculation:
wherein c, d represents the position of the current detected transient start point, s (c) represents the starting point amplitude, s (d) represents the end point amplitude, s 1 ,s 2 C is the amplitude value corresponding to the ideal transient starting and stopping point position c,d (u) is the log power cosine coefficient of the currently detected transient signal.
Further, in S4, constructing a radio frequency fingerprint feature vector based on the transformation result includes: and obtaining a maximum value, a minimum value, a first extreme value of a first-order difference quantity, a second-order central moment and a first-order origin moment of logarithmic envelope Fourier transformation of a transformation result to construct a radio frequency fingerprint feature vector.
Further, in S5, inputting the radio frequency fingerprint feature vector to a support vector machine for identity authentication of a device, including:
and performing dimension reduction on the radio frequency fingerprint feature vector by using principal component analysis, randomly dividing the dimension reduced features into a training set and a testing set according to a preset proportion, inputting the training set and the testing set into a Support Vector Machine (SVM) for classification, and comparing a label result output by the trained model with an actual label to obtain the classification accuracy.
The invention also discloses a radio frequency fingerprint identification system based on the iterative cosine spectrum transformation, which comprises:
a power envelope acquisition unit, configured to acquire a transmitter signal to be identified, and separate a power envelope and a carrier from the transmitter signal to be identified;
the primary transient interval acquisition unit is used for acquiring a primary transient interval according to a preset threshold after performing first-order derivation on the power envelope, and acquiring a transient signal in the primary transient interval;
the acquisition unit is used for establishing a floating interval at the starting and stopping positions of the transient signal based on the duration of the transient signal, establishing a fixed step length to set a sliding window to slide in the floating interval, continuously calculating the value of a cost function in the sliding process, and adjusting the sliding step length until the position which enables the cost function value to be minimum is found when the calculated cost function value is smaller than the cost function value of the transient signal, so that an optimal target interval is acquired;
the radio frequency fingerprint feature vector construction unit is used for constructing a radio frequency fingerprint feature vector based on a transformation result after carrying out fast Fourier transformation and discrete cosine transformation on signals in the optimal target interval in sequence;
and the identity authentication unit is used for inputting the radio frequency fingerprint feature vector into a support vector machine for identity authentication of equipment.
Compared with the prior art, the invention has the following advantages:
the method of the invention carries out deep analysis on transient power envelope characteristics, designs a three-phase down-conversion structure to realize turn-on power envelope and carrier classification, and can select the optimal target signal interval for radio frequency fingerprint extraction and identification no matter what modulation mode the received signal is, without other priori knowledge. The invention can effectively identify different devices and has an identification accuracy of 98 percent.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, and it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings may be obtained according to the drawings without inventive effort to a person skilled in the art.
FIG. 1 is a flowchart of a method for identifying a radio frequency fingerprint based on iterative cosine spectral transformation.
Fig. 2 is a flowchart illustrating an rf fingerprint identification implementation procedure in an embodiment.
FIG. 3 is a three-dimensional visualization of extracted features in an embodiment.
FIG. 4 shows the recognition accuracy achieved by the embodiment.
Detailed Description
In order that those skilled in the art will better understand the present invention, a technical solution in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in which it is apparent that the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present invention without making any inventive effort, shall fall within the scope of the present invention.
As shown in fig. 1, the invention provides a radio frequency fingerprint identification method based on iterative cosine spectrum transformation, which comprises the following steps:
s1, acquiring a transmitter signal to be identified, and separating a power envelope and a carrier from the transmitter signal to be identified.
Further, when the transmitter signal to be identified is acquired in this step, filtering is performed to remove spurious components that are not acquired during the acquisition process.
S2, after first-order derivation is carried out on the power envelope, a primary transient interval is obtained according to a preset threshold value, and a transient signal is obtained in the primary transient interval.
Specifically, through the step, firstly, transient rough detection is carried out, a first derivative is obtained on the power envelope, and a threshold value is set to obtain a transient interval detected in the first step.
Then intercepting the power envelope in turn-on logarithmic form in the interval, performing fast Fourier transform to the frequency domain, and continuing discrete cosine transform on the basis of the frequency domain.
S3, establishing a floating interval at a starting position and a stopping position of the transient signal based on the duration of the transient signal, establishing a fixed step length to set a sliding window to slide in the floating interval, continuously calculating the value of a cost function in the sliding process, and adjusting the sliding step length until a position which enables the cost function value to be minimum is found when the calculated cost function value is smaller than the cost function value of the transient signal, so that an optimal target interval is obtained.
S4, after the signal in the optimal target interval is subjected to fast Fourier transform and discrete cosine transform in sequence, a radio frequency fingerprint feature vector is constructed based on a transformation result.
S3 and S4, mainly used for transient detection optimization based on dynamic sliding window sequence search algorithm, extracting feature construction feature vector, specifically:
first, by designing a dynamic sliding window, the cost function is:
wherein c, d represents the position of the current transient starting point, s 1 ,s 2 For the amplitude value corresponding to the ideal transient starting and stopping point position, s (C) represents the starting amplitude, s (d) represents the ending amplitude, and C c,d (u) represents the target signal at [ c, d ]]Log cosine power spectrum in interval. Determining a more accurate transient signal interval position, and taking a signal interval corresponding to the minimum value of the cost function as an optimal target signal interval;
and then performing fast Fourier transform and discrete cosine transform processing by using an optimal target signal interval, and extracting seven-dimensional statistics of a maximum value, a minimum value, a first extreme value of a first-order difference quantity, a second extreme value of the first-order difference quantity, a second-order central moment of the first-order difference quantity, a first-order origin moment of logarithmic envelope Fourier transform and medium-high-dimensional energy of a logarithmic power cosine spectrum to construct a radio frequency fingerprint feature vector.
S5, inputting the radio frequency fingerprint feature vector to a support vector machine for identity authentication of equipment.
Specifically, the step uses principal component analysis to reduce the dimension of the obtained feature vector, randomly divides the feature after dimension reduction into a training set and a testing set according to a preset proportion, inputs the training set and the testing set into a Support Vector Machine (SVM) for classification, and compares a label result output by a trained model with an actual label to obtain the classification accuracy.
The following further describes the solution and effects of the present invention by means of specific application examples.
As shown in fig. 1, the radio frequency fingerprint identification method based on iterative cosine spectrum transformation provided in this embodiment includes:
1) The transmitter signal to be identified is acquired and the signal is subjected to power envelope and carrier separation.
In this embodiment, bluetooth wireless signals of 7 different devices are selected as target signals and numbered 1-7, and 150 pieces of data are collected by each device. Since bluetooth signals operate in ISM2400 band, the sampling frequency needs to be at least 4.8Gsps according to the nyquist sampling theorem. The data selected for this embodiment has a sampling rate of 5Gsps and a signal duration of about 10 mus. The real signal is directly captured by a high sampling rate oscilloscope and a low resolution analog-to-digital converter, and during the signal acquisition process, a band-pass filter is added in the preprocessing part to filter out spurious signals because the oscilloscope generates unnecessary spurious signals. Modeling the receiver signal as:
s IF (t)＝A env (t)·m IF (t)
wherein A is env And (t) is a power envelope model of the received signal, specifically expressed as:
m IF (t) is a carrier modulation signal, t represents time, t 0 Indicating the moment of transient initiation, t 1 Indicating the moment of transient end. Construction of s by designing a three-phase down-conversion structure IF The conjugate component of (t) completely eliminates the carrier component containing the modulation information through logarithmic operation, realizes the separation of the power envelopes and performs the energy normalization processing of the power envelopes, and the representation of the turn-on normalized logarithmic power envelopes is obtained as follows:
wherein A is s And (t) is the amplitude of the carrier modulated signal.
2) Performing logarithmic power cosine transform on the target signal;
in the step, firstly, transient rough detection is carried out, a first derivative is obtained for a power envelope, and the position of a midpoint of a transient part is obtained according to symmetry of transient envelope derivatives, so that a transient interval detected in the first step is obtained. The starting point of the transient interval is the starting time of the transmitter, and the end point is the time when the transmitting power reaches the rated power.
Carrying out logarithmic power cosine transform on the signals in the transient interval which is detected roughly to obtain a logarithmic power cosine coefficient:
wherein:
F ramp,1 (k)+F ramp,2 (k) Is f of ramp The result of performing a fast fourier transform in discrete form of (t). It should be noted that since the invention only deals with transient signals, the transient envelope amplitude is zero by default at time t=0, and thus the logarithmic form of the envelope can be expanded into the form of a power series. Wherein the denominator portion is in discrete form log (A s (t)) and the numerator portion is a power series expansion corresponding to the discrete form logarithmic power envelope. N is FFT point number, u represents the order of logarithmic cosine power coefficient, a s,q Coefficients of the expansion term for logarithmic form carrier power series, a env,p The coefficients of the power series expansion term are enveloped in logarithmic form, and n represents the number of sampling points of the discrete signal.
From observation, F ramp,1 (k) As a constant term, only F ramp,2 (k) The values of (C) vary, whereby it can be deduced that the values of the log-power cosine spectrum C (u) are uniquely related to the coefficients of the power series expansion of the power envelope only.
3) Transient detection optimization based on a dynamic sliding window sequence search algorithm, feature construction feature vector extraction comprises the following steps:
a. designing a cost function:
wherein c, d represents the position of the currently detected transient start point, s 1 ,s 2 The amplitude value corresponding to the ideal transient start-stop point position. First, the approximate position p, q of the transient signal obtained in step 2) yields a continuous sample length of the transient portion of w=p-q. Left and right floating sections with the size of w/8 are respectively arranged at the position p and the position q, and the step epsilon is firstly adopted in the floating sections 1 Setting a sliding window (5 sample points), obtaining the value of the cost function every time the sliding window is needed, and changing the sliding step length into epsilon when the current position cost function value is smaller than J (p, q) 2 (1 sample point), finding out the result which can minimize the cost function value is the optimal target interval.
b. And (3) performing fast Fourier transform and discrete cosine transform processing in the step (2) by using an optimal target signal interval, taking a logarithmic power cosine spectrum, and then obtaining statistics such as a maximum value, a minimum value, a first extreme value of a first-order difference component, a second-order central moment, a first-order origin moment of logarithmic envelope Fourier transform and the like by the logarithmic power cosine spectrum to construct a radio frequency fingerprint feature vector.
4) And inputting the feature vector into a support vector machine for identity authentication of the equipment.
Specifically, 150 samples per device were taken at 8:2 is divided into a training set and a testing set, the main component analysis is used for carrying out dimension reduction treatment on the obtained seven-dimensional vector, and the treatment flow of the main component analysis is as follows:
a. for feature set X ε R n*m Is de-averaged. For each feature x of the sample i Using the value x of the current feature i Subtracting the mean value of the feature in the sample set, namely:
b. calculating covariance matrix XX of sample T 。
c. And carrying out eigenvalue decomposition on the covariance matrix, and solving eigenvalues and eigenvectors of the covariance matrix.
d. And selecting the feature vector corresponding to the top k maximum feature values to be converted into P.
e.Y =px is the obtained k-dimensional feature matrix after the dimension reduction.
The feature vector after dimension reduction is used as input to a Support Vector Machine (SVM) to learn and train a model by using a linear kernel function, and the classification accuracy of the obtained model is 97.2% and 98.5% on a training set and a testing set respectively.
The invention also discloses a radio frequency fingerprint identification system based on the iterative cosine spectrum transformation, which comprises:
a power envelope acquisition unit, configured to acquire a transmitter signal to be identified, and separate a power envelope and a carrier from the transmitter signal to be identified;
the primary transient interval acquisition unit is used for acquiring a primary transient interval according to a preset threshold after performing first-order derivation on the power envelope, and acquiring a transient signal in the primary transient interval;
the acquisition unit is used for establishing a floating interval at the starting and stopping positions of the transient signal based on the duration of the transient signal, establishing a fixed step length to set a sliding window to slide in the floating interval, continuously calculating the value of a cost function in the sliding process, and adjusting the sliding step length until the position which enables the cost function value to be minimum is found when the calculated cost function value is smaller than the cost function value of the transient signal, so that an optimal target interval is acquired;
the radio frequency fingerprint feature vector construction unit is used for constructing a radio frequency fingerprint feature vector based on a transformation result after carrying out fast Fourier transformation and discrete cosine transformation on signals in the optimal target interval in sequence;
and the identity authentication unit is used for inputting the radio frequency fingerprint feature vector into a support vector machine for identity authentication of equipment.
For the embodiments of the present invention, since they correspond to those in the above embodiments, the description is relatively simple, and the relevant similarities will be found in the description of the above embodiments, and will not be described in detail herein.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and not for limiting the same; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some or all of the technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit of the invention.
Claims (5)
1. The radio frequency fingerprint identification method based on the iterative cosine spectrum transformation is characterized by comprising the following steps of:
s1, acquiring a transmitter signal to be identified, and separating a power envelope and a carrier from the transmitter signal to be identified;
s2, after first-order derivation is carried out on the power envelope, a primary transient interval is obtained according to a preset threshold value, and a transient signal is obtained in the primary transient interval;
s3, establishing a floating interval at a start position and a stop position of the transient signal based on the duration of the transient signal, establishing a fixed step length, setting a sliding window to slide in the floating interval, continuously calculating the value of a cost function in the sliding process, and adjusting the sliding step length until a position which enables the cost function value to be minimum is found when the calculated cost function value is smaller than the cost function value of the transient signal, thereby obtaining an optimal target interval, wherein the cost function is obtained according to the following calculation:
wherein the method comprises the steps ofC, d represents the position of the currently detected transient start-stop point, s (c) represents the start-point amplitude, s (d) represents the end-point amplitude, s 1 ,s 2 C is the amplitude value corresponding to the ideal transient starting and stopping point position c,d (u) is the log power cosine spectrum of the currently detected transient signal;
s4, after performing fast Fourier transform and discrete cosine transform on signals in an optimal target interval in sequence, constructing a radio frequency fingerprint feature vector based on a transformation result, wherein the method comprises the following steps: obtaining a maximum value, a minimum value, a first extreme value of a first-order difference quantity, a second-order central moment and a first-order origin moment of logarithmic envelope Fourier transform of a transformation result to construct a radio frequency fingerprint feature vector;
s5, inputting the radio frequency fingerprint feature vector to a support vector machine for identity authentication of equipment.
2. The method for identifying the radio frequency fingerprint based on the iterative cosine spectrum transformation according to claim 1, wherein after the transmitter signal to be identified is acquired in the step S1, filtering processing is further performed on the transmitter signal to be identified, so as to filter spurious components which are not acquired in the acquisition process.
3. The method for identifying the radio frequency fingerprint based on the iterative cosine spectrum transformation according to claim 1, wherein the step of establishing the floating interval at the start and stop positions of the transient signal based on the duration of the transient signal in the step S3 comprises the steps of:
a floating range with the size of w/8 is set at the starting and stopping positions of the primary transient interval, so that a floating interval is constructed, wherein w represents the duration of the transient signal.
4. The method for identifying the radio frequency fingerprint based on the iterative cosine spectrum transformation according to claim 1, wherein the step S5 of inputting the radio frequency fingerprint feature vector to a support vector machine for identity authentication of equipment comprises the following steps:
and performing dimension reduction on the radio frequency fingerprint feature vector by using principal component analysis, randomly dividing the dimension reduced features into a training set and a testing set according to a preset proportion, inputting the training set and the testing set into a Support Vector Machine (SVM) for classification, and comparing a label result output by the trained model with an actual label to obtain the classification accuracy.
5. An iterative cosine spectrum transformation-based radio frequency fingerprint identification system, comprising:
a power envelope acquisition unit, configured to acquire a transmitter signal to be identified, and separate a power envelope and a carrier from the transmitter signal to be identified;
the primary transient interval acquisition unit is used for acquiring a primary transient interval according to a preset threshold after performing first-order derivation on the power envelope, and acquiring a transient signal in the primary transient interval;
the acquisition unit is used for establishing a floating interval at the starting and stopping positions of the transient signal based on the duration of the transient signal, establishing a fixed step length to set a sliding window to slide in the floating interval, continuously calculating the value of a cost function in the sliding process, and adjusting the sliding step length until the position which enables the cost function value to be minimum is found when the calculated cost function value is smaller than the cost function value of the transient signal, so as to acquire an optimal target interval, wherein the cost function is obtained according to the following calculation:
wherein c, d represents the position of the current detected transient start point, s (c) represents the starting point amplitude, s (d) represents the end point amplitude, s 1 ,s 2 C is the amplitude value corresponding to the ideal transient starting and stopping point position c,d (u) is the log power cosine spectrum of the currently detected transient signal;
the radio frequency fingerprint feature vector construction unit is used for constructing a radio frequency fingerprint feature vector based on a transformation result after carrying out fast Fourier transformation and discrete cosine transformation on signals in an optimal target interval in sequence, and comprises the following steps: obtaining a maximum value, a minimum value, a first extreme value of a first-order difference quantity, a second-order central moment and a first-order origin moment of logarithmic envelope Fourier transform of a transformation result to construct a radio frequency fingerprint feature vector;
and the identity authentication unit is used for inputting the radio frequency fingerprint feature vector into a support vector machine for identity authentication of equipment.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210376325.4A CN114978582B (en) | 2022-04-11 | 2022-04-11 | Radio frequency fingerprint identification method and system based on iterative cosine spectrum transformation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210376325.4A CN114978582B (en) | 2022-04-11 | 2022-04-11 | Radio frequency fingerprint identification method and system based on iterative cosine spectrum transformation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114978582A CN114978582A (en) | 2022-08-30 |
| CN114978582B true CN114978582B (en) | 2023-08-04 |
Family
ID=82977997
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210376325.4A Active CN114978582B (en) | 2022-04-11 | 2022-04-11 | Radio frequency fingerprint identification method and system based on iterative cosine spectrum transformation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114978582B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106446877A (en) * | 2016-11-21 | 2017-02-22 | 电子科技大学 | Signal amplitude sequence based radio frequency fingerprint feature extraction and identification method |
| CN109598216A (en) * | 2018-11-23 | 2019-04-09 | 南方电网科学研究院有限责任公司 | A kind of radio-frequency fingerprint feature extracting method based on convolution |
| CN114282803A (en) * | 2021-12-23 | 2022-04-05 | 北京东土拓明科技有限公司 | Emergency disposal method and device, computing equipment and computer readable storage medium |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10834531B2 (en) * | 2005-12-15 | 2020-11-10 | Polte Corporation | Multi-path mitigation in rangefinding and tracking objects using reduced attenuation RF technology |
| US11176589B2 (en) * | 2018-04-10 | 2021-11-16 | Ebay Inc. | Dynamically generated machine learning models and visualization thereof |
-
2022
- 2022-04-11 CN CN202210376325.4A patent/CN114978582B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106446877A (en) * | 2016-11-21 | 2017-02-22 | 电子科技大学 | Signal amplitude sequence based radio frequency fingerprint feature extraction and identification method |
| CN109598216A (en) * | 2018-11-23 | 2019-04-09 | 南方电网科学研究院有限责任公司 | A kind of radio-frequency fingerprint feature extracting method based on convolution |
| CN114282803A (en) * | 2021-12-23 | 2022-04-05 | 北京东土拓明科技有限公司 | Emergency disposal method and device, computing equipment and computer readable storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114978582A (en) | 2022-08-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Köse et al. | RF fingerprinting of IoT devices based on transient energy spectrum | |
| CN105678273A (en) | Initial point detection algorithm of transient signal in radio frequency fingerprint identification technology | |
| CN107392123B (en) | Radio frequency fingerprint feature extraction and identification method based on coherent accumulation noise elimination | |
| CN108806718A (en) | Based on the audio authentication method to ENF phase spectrums and instantaneous frequency spectrum analysis | |
| CN108090462A (en) | A kind of Emitter Fingerprint feature extracting method based on box counting dimension | |
| CN105429719A (en) | Strong interference signal detection method based on power spectrum and multiple dimensioned wavelet transformation analysis | |
| Zhang et al. | Radio frequency fingerprint identification based on logarithmic power cosine spectrum | |
| CN111027614A (en) | Noise-enhanced radio frequency fingerprint identification method and device | |
| CN113343868A (en) | Radiation source individual identification method and device, terminal and storage medium | |
| Zhou et al. | A novel radio frequency fingerprint identification method using incremental learning | |
| CN114978582B (en) | Radio frequency fingerprint identification method and system based on iterative cosine spectrum transformation | |
| CN111245821B (en) | Radiation source identification method and device and radiation source identification model creation method and device | |
| Wang et al. | A new method of automatic modulation recognition based on dimension reduction | |
| CN109598216B (en) | Convolution-based radio frequency fingerprint feature extraction method | |
| Wang et al. | Specific emitter identification based on fractal and wavelet theories | |
| Ding et al. | Radio frequency fingerprint extraction based on singular values and singular vectors of time-frequency spectrum | |
| Liu et al. | RF fingerprint recognition based on spectrum waterfall diagram | |
| Liu et al. | Spectrum-based fingerprint extraction and identification method of 100m ethernet card | |
| CN116132991A (en) | Radio frequency fingerprint authentication method, device and storage medium of RKE system | |
| Wu et al. | Authorized and rogue LTE terminal identification using wavelet coefficient graph with auto-encoder | |
| CN115103365A (en) | Mobile communication terminal identification method based on polarization fingerprint | |
| CN111814703B (en) | HB-based signal joint feature extraction method under non-reconstruction condition | |
| CN113065385A (en) | Communication radiation source individual identification method based on axial integral bispectrum | |
| Wang et al. | Novel radio-frequency fingerprint recognition scheme using multiwavelets-based cyclic-spectrum graph analysis | |
| Zhang et al. | Research on Individual Identification of Wireless Devices Based on Signal's Energy Distribution |
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 |