CN115065973A - Convolutional neural network-based satellite measurement and control ground station identity recognition method - Google Patents
Convolutional neural network-based satellite measurement and control ground station identity recognition method Download PDFInfo
- Publication number
- CN115065973A CN115065973A CN202210372753.XA CN202210372753A CN115065973A CN 115065973 A CN115065973 A CN 115065973A CN 202210372753 A CN202210372753 A CN 202210372753A CN 115065973 A CN115065973 A CN 115065973A
- Authority
- CN
- China
- Prior art keywords
- signal
- neural network
- radio frequency
- target
- network model
- 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
- 238000000034 method Methods 0.000 title claims abstract description 68
- 238000005259 measurement Methods 0.000 title claims abstract description 30
- 238000013527 convolutional neural network Methods 0.000 title abstract description 6
- 238000003062 neural network model Methods 0.000 claims abstract description 67
- 238000007781 pre-processing Methods 0.000 claims abstract description 16
- 238000012545 processing Methods 0.000 claims description 39
- 238000012549 training Methods 0.000 claims description 21
- 230000008569 process Effects 0.000 claims description 19
- 238000006243 chemical reaction Methods 0.000 claims description 17
- 238000012360 testing method Methods 0.000 claims description 12
- 238000012795 verification Methods 0.000 claims description 12
- 238000004590 computer program Methods 0.000 claims description 8
- 238000001514 detection method Methods 0.000 claims description 7
- 238000010606 normalization Methods 0.000 claims description 6
- 238000000605 extraction Methods 0.000 claims description 4
- 230000009467 reduction Effects 0.000 claims description 4
- 230000004044 response Effects 0.000 claims description 3
- 230000011218 segmentation Effects 0.000 claims description 3
- 238000003860 storage Methods 0.000 claims description 3
- 238000005070 sampling Methods 0.000 description 32
- 238000010586 diagram Methods 0.000 description 11
- 238000013135 deep learning Methods 0.000 description 5
- 230000007613 environmental effect Effects 0.000 description 5
- 239000011159 matrix material Substances 0.000 description 5
- 238000004891 communication Methods 0.000 description 4
- 238000009825 accumulation Methods 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 240000007651 Rubus glaucus Species 0.000 description 2
- 235000011034 Rubus glaucus Nutrition 0.000 description 2
- 235000009122 Rubus idaeus Nutrition 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- QVFWZNCVPCJQOP-UHFFFAOYSA-N chloralodol Chemical compound CC(O)(C)CC(C)OC(O)C(Cl)(Cl)Cl QVFWZNCVPCJQOP-UHFFFAOYSA-N 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000010200 validation analysis Methods 0.000 description 2
- 108010003272 Hyaluronate lyase Proteins 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000005562 fading Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000012821 model calculation Methods 0.000 description 1
- 230000010355 oscillation Effects 0.000 description 1
- 230000001131 transforming effect Effects 0.000 description 1
Images
Classifications
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B7/00—Radio transmission systems, i.e. using radiation field
- H04B7/14—Relay systems
- H04B7/15—Active relay systems
- H04B7/185—Space-based or airborne stations; Stations for satellite systems
- H04B7/18578—Satellite systems for providing broadband data service to individual earth stations
- H04B7/18593—Arrangements for preventing unauthorised access or for providing user protection
Landscapes
- Engineering & Computer Science (AREA)
- Computer Security & Cryptography (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Signal Processing (AREA)
- Computer Networks & Wireless Communication (AREA)
- Theoretical Computer Science (AREA)
- Biophysics (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Mathematical Physics (AREA)
- Data Mining & Analysis (AREA)
- Biomedical Technology (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Astronomy & Astrophysics (AREA)
- Aviation & Aerospace Engineering (AREA)
- Monitoring And Testing Of Transmission In General (AREA)
Abstract
The embodiment of the invention discloses a convolutional neural network-based satellite measurement and control ground station identity recognition method, which comprises the steps of acquiring a radio frequency signal sent by equipment to be recognized, converting the radio frequency signal into a baseband signal, preprocessing the baseband signal to determine a target signal, and inputting the target signal into a pre-trained neural network model for recognition to determine a recognition result. Therefore, the radio frequency signal is converted into the baseband signal, and the baseband signal is identified according to the preprocessed baseband signal and the pre-trained neural network model, so that the key information of the radio frequency signal is kept, the accuracy of the identification result is improved, meanwhile, the model identification calculated amount can be reduced, and the identification efficiency is improved.
Description
Technical Field
The invention relates to the technical field of computers, in particular to a satellite measurement and control ground station identity recognition method based on a convolutional neural network.
Background
The radio frequency fingerprint identification technology extracts the radio frequency fingerprint of a signal to identify equipment by analyzing the radio frequency signal of the wireless communication equipment, so that the identity authentication of a physical layer is realized, and the safety of wireless communication is improved.
The traditional radio frequency fingerprint identification method realizes identity verification by artificially selecting the characteristics in a radio frequency signal and designing a classifier, but the characteristics selected by the method are greatly influenced by the channel environment, the identification method has poor applicability and the accuracy is easily influenced. In addition, the radio frequency identification method based on deep learning is superior to the traditional radio frequency fingerprint identification method in generality and identification accuracy, but the neural network parameters are numerous, the calculation amount is large, and the identification efficiency is low.
Disclosure of Invention
In view of this, the embodiment of the present invention provides a method for identifying an identity of a satellite measurement and control ground station based on a convolutional neural network, so as to improve accuracy and identification efficiency of an equipment identification result.
In a first aspect, an embodiment of the present invention provides an apparatus identification method, where the method includes:
acquiring a radio frequency signal sent by equipment to be identified;
converting the radio frequency signal into a baseband signal;
preprocessing the baseband signal to determine a target signal;
and inputting the target signal into a pre-trained neural network model for recognition so as to determine a recognition result.
Further, the converting the radio frequency signal into a baseband signal includes:
and performing down-conversion processing on the radio frequency signal, and determining the in-phase orthogonal signal after down-conversion processing as the baseband signal.
Further, the neural network model includes:
the characteristic unit is used for carrying out dimension reduction and characteristic extraction on the target signal;
the convolution unit comprises a first preset number of convolution layers and is used for performing convolution processing on the feature information extracted by the feature unit;
and the classification unit comprises a second preset number of full connection layers and is used for classifying the processed characteristic information so as to determine an identification result.
Further, the first preset quantity and the second preset quantity are determined based on a pre-training process corresponding to the neural network model.
Further, the baseband signals include an in-phase baseband signal and a quadrature baseband signal, and the pre-processing the baseband signals to determine the target signal includes:
determining a steady-state section signal of the baseband signal based on an energy detection method, wherein the steady-state section signal comprises an in-phase steady-state signal and a quadrature steady-state signal;
respectively carrying out normalization processing on the in-phase steady-state signal and the orthogonal steady-state signal;
carrying out data segmentation processing on the normalized in-phase steady-state signal and the normalized quadrature steady-state signal to determine an in-phase target sub-signal and a quadrature target sub-signal of at least one time period;
and respectively combining the in-phase target sub-signal and the orthogonal target sub-signal in each same time interval to determine a target signal in a two-dimensional data form.
Furthermore, the equipment to be identified is a measurement and control ground station, and the neural network model is deployed at a satellite terminal.
Further, the neural network model is determined based on the steps of:
acquiring original data;
determining a training set, a verification set and a test set according to the original data;
inputting the data in the training set into a preset neural network model for training, and determining at least one group of model parameters;
inputting the data corresponding to the verification set into the neural network models corresponding to each group of model parameters so as to determine target model parameters according to the identification results corresponding to each neural network model;
inputting data corresponding to the test set into a neural network model corresponding to the target model parameters, and determining the accuracy of the current identification result;
and determining the neural network model corresponding to the target model parameter as the trained neural network model in response to the accuracy of the current recognition result meeting a preset condition.
In a second aspect, an embodiment of the present invention provides an apparatus for identifying a device, where the apparatus includes:
the device comprises an acquisition unit, a processing unit and a processing unit, wherein the acquisition unit is used for acquiring a radio frequency signal sent by the device to be identified;
the conversion unit is used for converting the radio frequency signal into a baseband signal;
the preprocessing unit is used for preprocessing the baseband signal to determine a target signal;
and the recognition unit is used for inputting the target signal into a pre-trained neural network model for recognition so as to determine a recognition result.
In a third aspect, an embodiment of the present invention provides an electronic device, including a memory and a processor, where the memory is used to store one or more computer program instructions, where the one or more computer program instructions are executed by the processor to implement the method described in any one of the above.
In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, in which a computer program is stored, and the computer program, when executed by a processor, implements the method steps described in any one of the above.
According to the technical scheme, the radio frequency signal sent by the equipment to be identified is converted into the baseband signal for identification, the characteristic that the modulation domain feature of a signal physical layer can not be forged is utilized, the key information of the radio frequency signal is kept, meanwhile, the situation that the radio frequency signal is decoded, counterfeited and influenced by environmental factors in the identification process can be avoided, and the accuracy of the identification result is improved. Moreover, the baseband signal corresponding to the preprocessed radio-frequency signal is processed through the pre-trained neural network model, in the process of determining the identity of the equipment to be recognized, the recognition result can be determined without manually extracting features, and the recognition efficiency is higher.
Drawings
The above and other objects, features and advantages of the present invention will become more apparent from the following description of the embodiments of the present invention with reference to the accompanying drawings, in which:
FIG. 1 is a schematic diagram of a device identification method of an embodiment of the present invention;
FIG. 2 is a flow chart of determining a target signal according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of determining a target signal according to an embodiment of the present invention;
FIG. 4 is a schematic diagram of a neural network model structure of an embodiment of the present invention;
FIG. 5 is a flow chart of determining a neural network model for an embodiment of the present invention;
fig. 6 is a flowchart of an embodiment of a device identification method according to the present invention;
FIG. 7 is a schematic diagram of a device identification apparatus of an embodiment of the present invention;
fig. 8 is a schematic diagram of an electronic device of an embodiment of the invention.
Detailed Description
The present invention will be described below based on examples, but the present invention is not limited to only these examples. In the following detailed description of the present invention, certain specific details are set forth. It will be apparent to one skilled in the art that the present invention may be practiced without these specific details. Well-known methods, procedures, flows, components and circuits have not been described in detail so as not to obscure the present invention.
Further, those of ordinary skill in the art will appreciate that the drawings provided herein are for illustrative purposes and are not necessarily drawn to scale.
Unless the context clearly requires otherwise, throughout the description, the words "comprise", "comprising", and the like are to be construed in an inclusive sense as opposed to an exclusive or exhaustive sense; that is, what is meant is "including, but not limited to".
In the description of the present invention, it is to be understood that the terms "first," "second," and the like are used for descriptive purposes only and are not to be construed as indicating or implying relative importance. In addition, in the description of the present invention, "a plurality" means two or more unless otherwise specified.
The tolerance of the electronic device in the production process can cause the signals generated by different wireless communication devices to have slight difference, even the wireless communication devices of the same manufacturer and the same model have certain difference due to the tolerance effect and have the characteristic of difficult counterfeiting, and the difference forms the unique radio frequency fingerprint of the transmitter.
The traditional radio frequency fingerprint identification method at present needs manual feature selection and a classifier is designed, the accuracy of an identification result depends on expert field knowledge mastered by an operator, the intelligent degree is low, and the identification accuracy is poor. Moreover, the traditional radio frequency fingerprint identification method is greatly influenced by the channel environment, the characteristics need to be manually selected and the classifier needs to be designed specially aiming at specific environmental conditions in the actual application scene, and the effect is poor when the environment changes.
With the development of big data and deep learning, the radio frequency fingerprint identification method based on deep learning exceeds the traditional method based on artificial features in method universality and identification accuracy. At present, a radio frequency fingerprint identification method combining a constellation diagram and deep learning and a radio frequency fingerprint identification method combining a transform domain characteristic and deep learning have been researched and proposed, but the methods have the problems of numerous neural network parameters and large calculation amount, and the application of the methods is limited.
In view of this, the embodiment of the present invention provides a method for identifying an identity of a satellite measurement and control ground station based on a convolutional neural network, so as to improve the identification efficiency and accuracy of device identification.
In the following, the radio frequency fingerprint identification process of the measurement and control ground station is taken as an example for description, and it should be understood that the method in this embodiment may also be applied to other scenarios of performing identity identification through radio frequency signals, and this is not limited here.
Fig. 1 is a schematic diagram of a device identification method according to an embodiment of the present invention. As shown in fig. 1, the device identification method of the present embodiment includes the following steps.
In step S110, a radio frequency signal sent by the device to be identified is acquired.
In this embodiment, the device to be identified may be a measurement and control ground station, or may be other devices that transmit radio frequency signals to operate. Taking the measurement and control ground station as an example, the radio frequency signal sent by the equipment to be identified is acquired by acquiring the wireless radio frequency signal sent by the measurement and control ground station to the satellite terminal. Since the measurement and control ground station continuously transmits the wireless radio frequency signal to the satellite terminal, in order to improve the identification efficiency, the received continuous wireless signal is periodically sampled based on a sampling method in this embodiment, so as to obtain the radio frequency signal transmitted by the device to be identified.
It should be understood that in other alternative implementations, a wireless signal may be collected for a continuous period of time as a target radio frequency signal to perform subsequent processing.
Optionally, in this embodiment, the radio frequency signal sent by the measurement and control ground station is sampled at a predetermined time according to a preset sampling time, a plurality of sampling signals of the preset sampling points are sampled, and all the collected sampling signals are arranged according to a time sequence to form a time sequence, so as to obtain the radio frequency signal to be identified corresponding to the current measurement and control ground station.
Specifically, assuming that the preset sampling time is T and the preset number of sampling points is m, each sampling corresponds to a segment of signal X (i), where i ═ 1, 2, ·, m, it is determined that the radio frequency signal sent by the device to be identified is X ═ X (1), X (2), ·, X (m).
Further, in this embodiment, the duration corresponding to the preset sampling time and the number of preset sampling points are set and adjusted according to the actual usage scenario and the radio frequency fingerprint identity of the device to be identified. For example, when the radio frequency fingerprint identity of the device to be identified contains a large amount of information, the duration corresponding to the preset sampling time may be longer, and the number of the preset sampling points may be larger, so as to further improve the accuracy of the identification result.
In step S120, the rf signal is converted into a baseband signal.
In this embodiment, an IQ signal (In-phase Quadrature signal) is used as the baseband signal. Because the baseband signals carry modulation domain characteristics such as IQ imbalance, phase offset and the like, the characteristic information can still be kept under the condition of large environmental influence, so that the influence of environmental factors on the accuracy of the identification result is avoided, and the accuracy of the identification result is improved. Based on this, the present embodiment performs down-conversion processing on the radio frequency signal, and determines the in-phase and quadrature signals after down-conversion processing as baseband signals.
The Down-conversion process is also called Digital Down Converters (DDC), the Digital Down-converter mainly comprises a Numerically Controlled Oscillator (NCO), a mixer (mixer), a sampler (ADC), a filter (filter), and the like, and the Down-conversion function is realized by mixing an Intermediate Frequency (IF) Digital signal acquired by the AD with a local Digital intermediate frequency carrier signal generated by the Numerically Controlled Oscillator (NCO) in the DDC, and then obtaining a baseband signal through a low-pass filter. The basic principle of digital down conversion is the same as that of analog down conversion, namely, an input signal is multiplied by a local oscillation signal, a radio frequency signal is shifted to a middle frequency band through frequency mixing, and then ADC sampling is carried out. The core process is to mix the Intermediate Frequency (IF) digital signal collected by AD with the local digital intermediate frequency carrier signal generated by the Numerically Controlled Oscillator (NCO) in DDC, and then down-convert the intermediate frequency signal to baseband.
Further, in this embodiment, by performing down-conversion processing on the radio frequency signal, two paths of baseband signals including the in-phase baseband signal and the quadrature baseband signal are obtained at the same time. Therefore, the radio frequency signals sent by the equipment to be identified are converted into two paths of baseband signals, other conversion calculations can be reduced, the calculated amount in the subsequent signal identification process is reduced, and the identification efficiency is improved.
In step S130, the baseband signal is preprocessed to determine a target signal.
In this embodiment, the baseband signal includes an in-phase baseband signal and a quadrature baseband signal.
Optionally, as shown in fig. 2, the determining the target signal by preprocessing the baseband signal according to this embodiment includes the following steps.
In step S210, a steady-state segment signal of the baseband signal is determined based on an energy detection method. Wherein the steady state segment signal comprises an in-phase steady state signal and a quadrature steady state signal.
The energy detection method is a signal detection method, and comprises the steps of performing operation of performing modulo and quadratic operation on a time domain signal sampling value; or, the FFT (fast Fourier transform) is used for transforming the signal to a frequency domain, and then the sampled value of the frequency domain signal is subjected to modulus operation and then squared to obtain the energy accumulation value of the signal in a certain frequency band range. And comparing the energy accumulation value with a preset threshold value to judge and determine a comparison result, and determining the existence state of the signal according to the comparison result. In this embodiment, if the energy accumulation value of the signal is higher than the preset threshold, it indicates that the signal is in the steady-state section.
In step S220, normalization processing is performed on the in-phase steady-state signal and the quadrature steady-state signal, respectively.
The normalization process has two forms, one is to change a number to a decimal between (0, 1), and the other is to change a dimensional expression to a dimensionless expression. The data are mapped into the range of 0-1 for processing through normalization processing, so that the digital signal processing process is more convenient and faster.
In step S230, the normalized in-phase steady-state signal and the normalized quadrature steady-state signal are subjected to data slicing processing, and an in-phase target sub-signal and a quadrature target sub-signal of at least one time period are determined.
In this embodiment, the in-phase steady-state signal and the quadrature steady-state signal are respectively intercepted at intervals of a certain number of sampling points to realize the segmentation processing, and the in-phase target sub-signal and the quadrature target sub-signal in at least one time period are determined.
In step S240, the in-phase target sub-signal and the quadrature target sub-signal of each same time period are respectively combined to determine a target signal in the form of two-dimensional data.
In this embodiment, the in-phase target sub-signal and the quadrature target sub-signal after the splitting processing both correspond to one-dimensional data, data in a two-dimensional matrix form is formed by combining the in-phase target sub-signal and the quadrature target sub-signal at the same time period, and the combined two-dimensional matrix data is determined as the target signal after the preprocessing.
Fig. 3 is a schematic diagram of determining a target signal according to an embodiment of the present invention. As shown in fig. 3, the I-path signal is an in-phase signal, and the Q-path signal is a quadrature signal. In this embodiment, the in-phase steady-state signal and the quadrature steady-state signal are intercepted every M sampling points, so as to obtain the segmented in-phase target sub-signal and quadrature target sub-signal. And then combining the signals corresponding to the same time period in the in-phase target sub-signal and the orthogonal target sub-signal to determine a target signal in a two-dimensional data form.
For convenience of understanding, it is assumed that the in-phase steady-state signal and the quadrature steady-state signal are truncated every M sampling points, and each of the sliced in-phase target sub-signal and quadrature target sub-signal includes 5 segments of signals, and each segment of signals includes M sampling signals, as shown in fig. 3. After combining the signals corresponding to the same period in the in-phase target sub-signal and the quadrature target sub-signal, 5 target signals in the form of two-dimensional data are determined, including the target signals S1, S2, S3, S4, and S5.
Optionally, in this embodiment, after the in-phase target sub-signal and the orthogonal target sub-signal are combined, MATLAB or Python software is used to perform data enhancement on the combined signal data, so as to increase the influence of multiple effects, such as doppler shift, fading, and the like, specific to the measurement and control signal, and determine the enhanced signal as the target signal. Therefore, the effective characteristics in the target signal corresponding to the equipment to be recognized are more obvious and easy to recognize by adopting a data enhancement method, and the accuracy of the recognition result can be further improved.
In step S140, the target signal is input into a neural network model trained in advance for recognition, so as to determine a recognition result. And the identification result is used for representing the identity of the equipment to be identified.
In this embodiment, the neural network model is deployed on the satellite terminal. The recognition result may be a label for characterizing the identity of the device to be recognized. Therefore, the satellite terminal receives the wireless radio frequency signal sent by the equipment to be identified, the radio frequency signal sent by the equipment to be identified is obtained according to the received wireless radio frequency signal, the radio frequency signal is converted into a baseband signal, the baseband signal is preprocessed, a target signal is determined, the target signal is input into a pre-trained neural network model to be identified, and the identity of the equipment to be identified is further determined.
Optionally, the satellite terminal in this embodiment may only receive the wireless radio frequency signal sent by the legal measurement and control ground station, and identify the received radio frequency signal through the above processing procedure, so as to determine the identity of the device to be identified.
Alternatively, the satellite terminal in this embodiment may receive radio frequency signals sent by various types of measurement and control ground stations, such as a legal measurement and control ground station and an illegal measurement and control ground station. Therefore, after the identity of the device to be recognized is determined, the recognition result is also judged in the embodiment. Specifically, the identification result output by the neural network model is matched with a legal ground station radio frequency fingerprint database which is arranged on the satellite terminal in advance, and when the legal ground station radio frequency fingerprint database comprises the radio frequency fingerprint corresponding to the measurement and control ground station represented by the identification result, the measurement and control ground station corresponding to the identification result is determined to be the legal ground station. Therefore, the identity legality of the measurement and control ground station can be determined by matching the identification result with the radio frequency fingerprint database of the legal ground station.
According to the technical scheme, the radio frequency signal sent by the equipment to be identified is converted into the baseband signal for identification, the characteristic that the modulation domain feature of a signal physical layer can not be forged is utilized, the situation that the radio frequency signal is decoded, counterfeited and influenced by environmental factors in the identification process can be avoided, and the accuracy of the identification result is improved. Moreover, the baseband signal corresponding to the preprocessed radio-frequency signal is processed through the pre-trained neural network model, in the process of determining the identity of the equipment to be recognized, the recognition result can be determined without manually extracting features, and the recognition efficiency is higher. In addition, by optimizing the model structure and model parameters of the preset neural network, the calculated amount in the identification process can be reduced, the deployment of the neural network model is further facilitated, and the identity identification efficiency is improved.
Fig. 4 is a schematic diagram of a neural network model structure according to an embodiment of the present invention. As shown in fig. 4, the neural network model structure in the present embodiment includes a feature unit 1, a convolution unit 2, and a classification unit 3. The feature unit 1 is used for performing dimension reduction and feature extraction on the target signal. The convolution unit 2 includes a first preset number of convolution layers for performing convolution processing on the feature information extracted by the feature unit. The classification unit 3 includes a second preset number of full connection layers, and is configured to classify the processed feature information to determine an identification result. The identification result is used for representing the identity of the equipment to be identified, and the identity of the equipment to be identified can be represented by identification in the form of a label and the like.
Optionally, the feature unit in this embodiment employs a Conv (1, 2) convolutional layer. Since the target signal includes two paths of information, i.e., an in-phase target sub-signal and an orthogonal target sub-signal, the input format of the neural network model in this embodiment is N × 2. The feature unit 1 receives a target signal corresponding to the device to be identified in the form of N × 2, and performs reduction and extraction of IQ correlation features on data corresponding to the target signal by using a convolution layer with a convolution kernel size of (1, 2). The IQ correlation characteristics include the phase relationship and amplitude relationship between the I-path signal and the Q-path signal in the target signal. Therefore, after the target signal is processed by the feature unit, different local features in the target signal can be conveniently extracted through the convolution layers with the first preset number, all the local features are changed into global features through the full-connection layers with the second preset number, and then the identification result is determined according to the global features.
Optionally, the first preset number and the second preset number in this embodiment are determined based on a pre-training process corresponding to the neural network model. Further, in the embodiment, the parameters in the model are tested by using a control variable method, and the optimal parameters are determined as much as possible while fully considering the relationship among the model parameters, the recognition accuracy and the robustness, so that the accuracy, the recognition efficiency and the recognition stability of the output result of the neural network model are improved.
Specifically, the specific numerical values of the first preset quantity and the second preset quantity in this embodiment are respectively corresponding minimum numerical values when the accuracy and the efficiency of the recognition result are considered. For example, if the accuracy and the efficiency of the recognition result of the output result of the feature unit after being processed by the X convolutional layers and the Y full link layers can reach preset values, or the overall performance can reach global optimum while considering the accuracy, the efficiency and the network robustness of the recognition result, the first preset number is X, and the second preset number is Y. Therefore, the number of convolution layers in the convolution unit and the number of full connection layers in the classification unit are determined through the process, the model parameters in the neural network model can be reduced, and the model calculation amount is reduced while the characteristic nonlinear mapping capability is ensured.
Further, in the embodiment, the convolution layer in the convolution unit adopts a small convolution kernel smaller than the preset convolution kernel, so as to further reduce the network parameter in the neural network model, and further improve the recognition efficiency. And adjusting the specific numerical value of the preset convolution kernel according to the actual use scene requirement.
Further, as shown in fig. 5, the present embodiment determines the neural network model by the following steps.
In step S310, raw data is acquired.
In this embodiment, the satellite terminal receives radio frequency signals sent by a plurality of measurement and control ground stations, and obtains original data by sampling each radio frequency signal, where the original data includes a plurality of sample data sets, and one measurement and control ground station corresponds to one sample data set.
Assuming that there are n measurement and control ground stations, when sampling a radio frequency signal, sampling every time T, and sampling m times, where original data a ═ a1, a2, A3, ·, An ], a sample data set Ai ═ x (1), x (2), ·, x (m) corresponding to each measurement and control ground station, where i ═ 1, 2, ·, n.
In step S320, a training set, a validation set, and a test set are determined from the raw data.
In this embodiment, after the digital receiver of the satellite terminal obtains the original data, the digital receiver first converts the radio frequency signals in each sample into baseband signals through down-conversion processing, and then performs preprocessing on the baseband signals corresponding to each sample to determine corresponding target signals. And finally, dividing all target signals to respectively determine a training set, a verification set and a test set.
Optionally, when preprocessing each baseband signal, the embodiment determines an in-phase steady-state signal and an orthogonal steady-state signal in a corresponding baseband signal based on an energy detection method, normalizes the in-phase steady-state signal and the orthogonal steady-state signal respectively, then performs slicing on the normalized in-phase steady-state signal and the normalized orthogonal steady-state signal based on a preset number of sampling points, determines an in-phase target sub-signal and an in-phase target sub-signal in each time period, and finally combines the in-phase target sub-signal and the orthogonal target sub-signal in each same time period respectively to determine a target signal in a two-dimensional data form.
Optionally, in this embodiment, after the in-phase target sub-signal and the quadrature target sub-signal are combined into a two-dimensional matrix, a sample label is marked on the two-dimensional matrix, and the two-dimensional matrix is determined as the target signal of the sample. Therefore, in the embodiment, the corresponding relation between the target signal and the corresponding device is established in a labeling mode, so that the model parameters are adjusted based on the output result of the neural network model and the label information in the training process.
In step S330, the data in the training set is input into a preset neural network model for training, and at least one set of model parameters is determined.
In step S340, the data corresponding to the verification set is input into the neural network models corresponding to the sets of model parameters, so as to determine the target model parameters according to the recognition results corresponding to the neural network models.
In this embodiment, if an error between an identification result output after the verification set input data is input to the neural network model and output data corresponding to the verification set input data is within a preset range, it is determined that the neural network model under the current model parameters is successfully verified. And after the verification is successful, determining the current model parameters as the target model parameters.
In step S350, the data corresponding to the test set is input into the neural network model corresponding to the target model parameter, and the accuracy of the current recognition result is determined.
It should be understood that, in the embodiment, when the neural network model is trained through the corresponding data in the training set, a group of model parameters can be obtained through one round of training, the target model parameters are determined through the output result of the neural network model corresponding to each group of model parameters after the corresponding data in the test set is input, and then the accuracy of the identification result of the neural network model corresponding to each group of target model parameters is determined through the verification set.
Optionally, in this embodiment, all target model parameter combinations may be obtained by training first, and then each set of target model parameters is verified respectively; or training to obtain a group of target model parameter combinations, determining the identification accuracy of the group of target model parameters, continuing training to obtain other target model parameter combinations, and respectively determining the identification accuracy corresponding to each group of target model parameters.
In step S360, in response to that the accuracy of the current recognition result satisfies the preset condition, the neural network model corresponding to the target model parameter is determined as the trained neural network model.
Optionally, in this embodiment, a target model parameter combination with the highest recognition result accuracy is determined as a preset model parameter, the preset model parameter is applied to a preset neural network model, and then the trained neural network model is determined, and the radio frequency signal of the device to be recognized is recognized through the trained neural network model, so as to determine the identity of the device to be recognized.
Optionally, in this embodiment, a target model parameter combination with an accuracy rate of the recognition result greater than a preset accuracy rate may also be determined as the preset model parameter. When a group of preset model parameters exist, the preset model parameters are applied to a preset neural network model so as to determine the trained neural network model, and the identity of the equipment to be identified is determined through the trained neural network model so as to improve the accuracy of identity identification; when a plurality of groups of target model parameters with accuracy rates meeting preset conditions exist, a group of model parameters with smaller model parameters, higher accuracy rates and better model robustness are preferentially selected as finally determined preset model parameters, and the preset model parameters are applied to a preset neural network model to identify the equipment to be identified, so that the actual application performance of the identity identification method is improved while the identification accuracy rate is ensured.
Fig. 6 is a flowchart of an apparatus identification method according to an embodiment of the present invention. As shown in fig. 6, the present embodiment determines the identification result of the device to be identified by the following steps.
In step S410, raw data is acquired.
In step S420, a training set, a validation set, and a test set are determined from the raw data.
In step S430, a preset neural network model is trained, verified and tested based on the training set, the verification set and the test set, and the trained neural network model is determined.
In step S440, a radio frequency signal of the device to be identified is received and converted into a baseband signal.
In step S450, the baseband signal is preprocessed to determine a target signal.
In step S460, the target signal is input to the trained neural network model for recognition, and a recognition result is determined.
In step S470, the identification result is determined based on the valid ground station rf fingerprint library, and the validity of the device to be identified is determined.
It should be noted that, the specific implementation method of each step in this embodiment has been described above, and is not described herein again.
Fig. 7 is a schematic diagram of a device identification apparatus according to an embodiment of the present invention. As shown in fig. 7, the device identifying apparatus of the present embodiment includes an acquiring unit 10, a converting unit 20, a preprocessing unit 30, and an identifying unit 40. The acquiring unit 10 is configured to acquire a radio frequency signal sent by a device to be identified. The conversion unit 20 is configured to convert the radio frequency signal into a baseband signal. The pre-processing unit 30 is configured to pre-process the baseband signal to determine a target signal. The recognition unit 40 is configured to input the target signal into a pre-trained neural network model for recognition, so as to determine a recognition result.
Alternatively, the obtaining unit 10 and the converting unit 20 in this embodiment may be embedded in a receiver on a satellite terminal, so as to receive the radio frequency signal sent by the device to be identified through the receiver, and determine the identification result of the device to be identified according to the radio frequency signal sent by the device to be identified. Further, the receiver architecture in this embodiment adopts an orthogonal sampling zero intermediate frequency receiver, receives a radio frequency signal sent by a device to be identified through the orthogonal sampling zero intermediate frequency receiver, samples the radio frequency signal according to a preset sampling duration and a sampling point number, determines the sampled signal obtained after sampling as the radio frequency signal to be identified, further performs down-conversion processing on the sampled radio frequency signal through an internal structure of the receiver, converts the radio frequency signal sampled by the device to be identified into a baseband signal, so as to identify the device to be identified based on the baseband signal subsequently, and further determines an identity identification result of the device to be identified.
Optionally, the preprocessing unit 30 in this embodiment is specifically configured to determine a steady-state segment signal of the baseband signal based on an energy detection method, perform normalization processing on the in-phase steady-state signal and the quadrature steady-state signal respectively, perform data slicing processing on the normalized in-phase steady-state signal and quadrature steady-state signal, determine an in-phase target sub-signal and a quadrature target sub-signal in at least one time period, combine the in-phase target sub-signal and the quadrature target sub-signal in each same time period, and determine a target signal in a two-dimensional data form. Wherein the steady state segment signal comprises an in-phase steady state signal and a quadrature steady state signal.
Optionally, the identifying unit 40 in this embodiment is further configured to determine the identification result. Specifically, the identification result output by the neural network model is matched with a legal ground station radio frequency fingerprint database which is arranged on the satellite terminal in advance, and when the legal ground station radio frequency fingerprint database comprises the radio frequency fingerprint corresponding to the measurement and control ground station represented by the identification result, the measurement and control ground station corresponding to the identification result is determined to be the legal ground station. Therefore, the identity legality of the measurement and control ground station can be determined by matching the identification result with the radio frequency fingerprint database of the legal ground station.
Fig. 8 is a schematic diagram of an electronic device of an embodiment of the invention. As shown in fig. 8, the electronic device of the present embodiment includes a receiving end 41 and a processing system 42. Wherein, the receiving end 41 is configured to receive a wireless radio frequency signal transmitted by the device to be identified. The processing system 42 includes a receiver 421 and a processor 422. The receiver 421 is configured to, after receiving the radio frequency signal sent by the device to be identified, sample the radio frequency signal sent by the device to be identified based on a preset sampling duration and a preset number of sampling points, and determine a sampled radio frequency signal. The processor 422 is configured to receive the sampled radio frequency signal sampled by the receiver 421, and process and identify the sampled radio frequency signal as a radio frequency signal sent by the device to be identified, thereby determining an identification result of the device to be identified.
Optionally, the receiving end in this embodiment adopts an RTL-SDR including an RTL2832U chip, the sampling precision is 7bit, and receives a radio frequency signal to be sent by the video device through an omnidirectional antenna with a gain of 3 dB. The processing unit adopts raspberry group 4B and is connected with the receiving end through a USB interface. Furthermore, application software is arranged in the processing unit, and the application software adopts a software development tool suite of the GNU Radio open source. The software version was 3.8. The software sets the frequency point of the receiving end to be 433MHz, and the sampling rate to be 2.0Sample/s, which is 4 times of the maximum bandwidth of the signal.
Further, in the test stage, a navigation satellite signal (i.e. a radio frequency signal) is simulated and transmitted by adopting the LoRa wireless transmission module as a signal transmitting terminal (i.e. a device to be identified). The configuration mode that the frequency point is 433MHz, the spreading factor is 7, the signal bandwidth is 500kHz, the transmitting power is 11dBm, the antenna gain is 3dB, the data transmitting content is filled with random numbers, and the baud rate is 9600 is adopted for transmitting the wireless radio frequency signal. Therefore, the identification of the equipment to be identified is realized by the LoRa wireless transmitting module, the RTL-SDR and the raspberry group 4B component radio frequency fingerprint identification real-time processing system and the execution of the processing process, so that the identification accuracy is high, and the identification efficiency is high.
It should be noted that the receiving end and the processing unit in this embodiment are only examples in a specific test or use scenario, and the specific receiving end and the processing unit may be set to implement the device identification method, which is not limited herein.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (10)
1. A method for device identification, the method comprising:
acquiring a radio frequency signal sent by equipment to be identified;
converting the radio frequency signal into a baseband signal;
preprocessing the baseband signal to determine a target signal;
and inputting the target signal into a pre-trained neural network model for recognition so as to determine a recognition result.
2. The method of claim 1, wherein converting the radio frequency signal to a baseband signal comprises:
and performing down-conversion processing on the radio frequency signal, and determining the in-phase orthogonal signal after down-conversion processing as the baseband signal.
3. The method of claim 1, wherein the neural network model comprises:
the characteristic unit is used for carrying out dimension reduction and characteristic extraction on the target signal;
the convolution unit comprises a first preset number of convolution layers and is used for performing convolution processing on the feature information extracted by the feature unit;
and the classification unit comprises a second preset number of full connection layers and is used for classifying the processed characteristic information so as to determine an identification result.
4. The method of claim 3, wherein the first predetermined number and the second predetermined number are determined based on a pre-training process corresponding to the neural network model.
5. The method of claim 1, wherein the baseband signals comprise an in-phase baseband signal and a quadrature baseband signal, and wherein pre-processing the baseband signals to determine a target signal comprises:
determining a steady-state segment signal of the baseband signal based on an energy detection method, wherein the steady-state segment signal comprises an in-phase steady-state signal and a quadrature steady-state signal;
respectively carrying out normalization processing on the in-phase steady-state signal and the orthogonal steady-state signal;
carrying out data segmentation processing on the normalized in-phase steady-state signal and the normalized quadrature steady-state signal to determine an in-phase target sub-signal and a quadrature target sub-signal of at least one time period;
and respectively combining the in-phase target sub-signal and the orthogonal target sub-signal in each same time interval to determine a target signal in a two-dimensional data form.
6. The method according to claim 1, wherein the device to be identified is a measurement and control ground station, and the neural network model is deployed in a satellite terminal.
7. The method of claim 1, wherein the neural network model is determined based on the steps of:
acquiring original data;
determining a training set, a verification set and a test set according to the original data;
inputting the data in the training set into a preset neural network model for training, and determining at least one group of model parameters;
inputting the data corresponding to the verification set into the neural network models corresponding to each group of model parameters so as to determine target model parameters according to the identification results corresponding to each neural network model;
inputting data corresponding to the test set into a neural network model corresponding to the target model parameters, and determining the accuracy of the current identification result;
and determining the neural network model corresponding to the target model parameter as the trained neural network model in response to the accuracy of the current recognition result meeting a preset condition.
8. An apparatus for device identification, the apparatus comprising:
the device comprises an acquisition unit, a processing unit and a processing unit, wherein the acquisition unit is used for acquiring a radio frequency signal sent by the device to be identified;
the conversion unit is used for converting the radio frequency signal into a baseband signal;
the preprocessing unit is used for preprocessing the baseband signal to determine a target signal;
and the recognition unit is used for inputting the target signal into a pre-trained neural network model for recognition so as to determine a recognition result.
9. An electronic device comprising a memory and a processor, wherein the memory is configured to store one or more computer program instructions, wherein the one or more computer program instructions are executed by the processor to implement the method of any of claims 1-7.
10. A computer-readable storage medium, characterized in that a computer program is stored in the computer-readable storage medium, which computer program, when being executed by a processor, carries out the method steps of any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210372753.XA CN115065973A (en) | 2022-04-11 | 2022-04-11 | Convolutional neural network-based satellite measurement and control ground station identity recognition method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210372753.XA CN115065973A (en) | 2022-04-11 | 2022-04-11 | Convolutional neural network-based satellite measurement and control ground station identity recognition method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115065973A true CN115065973A (en) | 2022-09-16 |
Family
ID=83197351
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210372753.XA Pending CN115065973A (en) | 2022-04-11 | 2022-04-11 | Convolutional neural network-based satellite measurement and control ground station identity recognition method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115065973A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116776227A (en) * | 2023-08-04 | 2023-09-19 | 中国人民解放军战略支援部队航天工程大学 | Satellite identity recognition method and device based on feature fusion |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109919015A (en) * | 2019-01-28 | 2019-06-21 | 东南大学 | A kind of radio-frequency fingerprint extraction and recognition methods based on more sampling convolutional neural networks |
| CN113343874A (en) * | 2021-06-18 | 2021-09-03 | 上海电机学院 | Large-scale radio signal identification method based on deep convolutional neural network |
| KR102347174B1 (en) * | 2021-07-21 | 2022-01-03 | 국방과학연구소 | Ensemble based radio frequency fingerprinting apparatus and method of identifying emitter using the same |
-
2022
- 2022-04-11 CN CN202210372753.XA patent/CN115065973A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109919015A (en) * | 2019-01-28 | 2019-06-21 | 东南大学 | A kind of radio-frequency fingerprint extraction and recognition methods based on more sampling convolutional neural networks |
| CN113343874A (en) * | 2021-06-18 | 2021-09-03 | 上海电机学院 | Large-scale radio signal identification method based on deep convolutional neural network |
| KR102347174B1 (en) * | 2021-07-21 | 2022-01-03 | 국방과학연구소 | Ensemble based radio frequency fingerprinting apparatus and method of identifying emitter using the same |
Non-Patent Citations (2)
| Title |
|---|
| JIABAO YU; AIQUN HU; GUYUE LI; LINNING PENG: ""A_Robust_RF_Fingerprinting_Approach_Using_Multisampling_Convolutional_Neural_Network"", 《 IEEE INTERNET OF THINGS JOURNAL》, 16 April 2019 (2019-04-16) * |
| 李古月;俞佳宝;胡爱群;: "基于设备与信道特征的物理层安全方法", 密码学报, no. 02, 15 April 2020 (2020-04-15) * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116776227A (en) * | 2023-08-04 | 2023-09-19 | 中国人民解放军战略支援部队航天工程大学 | Satellite identity recognition method and device based on feature fusion |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10985955B2 (en) | Method for automatically identifying modulation mode for digital communication signal | |
| CN108764077B (en) | Digital signal modulation classification method based on convolutional neural network | |
| CN110099019B (en) | LoRa modulation signal detection method based on deep learning | |
| EP1750467A1 (en) | A method and system for optimizing the use of the radio spectrum and computer program product thereof | |
| CN102780534B (en) | Universal electromagnetic signal automatic searching method | |
| Nafkha et al. | Experimental spectrum sensing measurements using USRP Software Radio platform and GNU-radio | |
| CN104378128B (en) | The method that wireless receiving platform and the Radio frequency interference generated for platform mitigate | |
| CN106357575A (en) | Multi-parameter jointly-estimated interference type identification method | |
| CN104796366A (en) | Communication signal system identification system and method | |
| CN115065973A (en) | Convolutional neural network-based satellite measurement and control ground station identity recognition method | |
| CN101977091B (en) | Method and system for monitoring electromagnetic spectrum | |
| CN102668658A (en) | Autocorrelation-based spectrum sensing for fm signals | |
| CN114584227B (en) | Automatic burst signal detection method | |
| CN113347637A (en) | Embedded wireless device based RF fingerprint identification method and device | |
| CN111624632A (en) | Satellite navigation signal capturing method and device | |
| CN116383603A (en) | Mixed electromagnetic threat system identification method and system based on deep learning under strong countermeasure environment | |
| CN113242201B (en) | Wireless signal enhanced demodulation method and system based on generation classification network | |
| CN110086554A (en) | A kind of signal recognition method based on frequency spectrum perception | |
| Sadkhan | A proposed digital modulated signal identification based on pattern recognition | |
| CN101807961B (en) | Method for realizing spectrum sensing based on bi-spectrum diagonal slice | |
| CN114006798A (en) | Signal processing method and device, electronic equipment and storage medium | |
| Lijun et al. | A signal demodulation algorithm based on generative adversarial networks | |
| CN110247868A (en) | A kind of blind demodulation method of satellite-signal, device and electronic equipment | |
| Ren et al. | Deep Learning Based Identification Method for Signal-Level Wireless Protocol | |
| Uppal et al. | Rich feature deep learning classifier for multiple simultaneous radio signals |
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 |