CN115494526A - GNSS deception jamming detection method and device, electronic equipment and storage medium - Google Patents

GNSS deception jamming detection method and device, electronic equipment and storage medium Download PDF

Info

Publication number
CN115494526A
CN115494526A CN202211038322.6A CN202211038322A CN115494526A CN 115494526 A CN115494526 A CN 115494526A CN 202211038322 A CN202211038322 A CN 202211038322A CN 115494526 A CN115494526 A CN 115494526A
Authority
CN
China
Prior art keywords
data
support vector
vector machine
machine model
gnss
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
Application number
CN202211038322.6A
Other languages
Chinese (zh)
Inventor
朱祥维
陈正坤
李静
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Sun Yat Sen University
Original Assignee
Sun Yat Sen University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Sun Yat Sen University filed Critical Sun Yat Sen University
Priority to CN202211038322.6A priority Critical patent/CN115494526A/en
Publication of CN115494526A publication Critical patent/CN115494526A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • GPHYSICS
    • G01MEASURING; TESTING
    • G01SRADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S19/00Satellite radio beacon positioning systems; Determining position, velocity or attitude using signals transmitted by such systems
    • G01S19/01Satellite radio beacon positioning systems transmitting time-stamped messages, e.g. GPS [Global Positioning System], GLONASS [Global Orbiting Navigation Satellite System] or GALILEO
    • G01S19/13Receivers
    • G01S19/21Interference related issues ; Issues related to cross-correlation, spoofing or other methods of denial of service
    • G01S19/215Interference related issues ; Issues related to cross-correlation, spoofing or other methods of denial of service issues related to spoofing

Landscapes

  • Engineering & Computer Science (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Radar Systems Or Details Thereof (AREA)

Abstract

The application discloses a GNSS deception jamming detection method and device, electronic equipment and a storage medium, and relates to the technical field of GNSS satellite signals. Reading a collected signal of a GNSS deception scene, and obtaining original data of different stages from the collected signal; extracting a plurality of characteristic parameters from original data in different stages, making the characteristic parameters into a data set, and dividing the data set into training data and test data according to a certain proportion; constructing a support vector machine model, and training the support vector machine model by using training data to obtain an optimal support vector machine model under a constraint condition as a classifier for deception detection; and inputting the test data into the classifier to perform deception signal detection to obtain a detection result. The method and the device have the advantages that the detection process is integrated with the advantages of multiple characteristic parameters in different stages, better detection performance is obtained, and meanwhile applicability is improved.

Description

GNSS deception jamming detection method and device, electronic equipment and storage medium
Technical Field
The present disclosure relates to the field of GNSS satellite signal technologies, and in particular, to a GNSS deception jamming detection method and apparatus, an electronic device, and a storage medium.
Background
Since a Global Navigation Satellite System (GNSS) is susceptible to deception jamming, the wide application of the GNSS in national defense security and economic construction is limited, and how to quickly and accurately detect the GNSS deception jamming becomes important. In recent years, many scholars analyze the influence of spoofed signals on different stages of a satellite navigation receiver and use various information, and most of the scholars adopt a single-parameter detection method such as signal power, signal quality and the like to detect whether the spoofed signals invade. The single parameter detection method is to identify whether a deception signal exists according to the change of a certain parameter before and after deception, and the method of signal quality monitoring and signal power monitoring is commonly used, and mainly utilizes the characteristics of related peak distortion and abnormal signal power change caused by deception signal intrusion. However, the single-parameter detection method only estimates the change of a certain characteristic of the receiver caused by the deception interference, cannot cope with a complicated and changeable deception interference scene, and has poor scene practicability.
Disclosure of Invention
Therefore, embodiments of the present application provide a GNSS deception jamming detection method, apparatus, electronic device, and storage medium, which can solve the technical problem that a complex and variable deception jamming scene cannot be handled through single-parameter detection, and the specific technical solution content is as follows:
in a first aspect, an embodiment of the present application provides a GNSS spoofing interference detection method, including:
reading an acquisition signal of a GNSS deception scene, and acquiring original data of different stages from the acquisition signal;
extracting a plurality of characteristic parameters from the original data of different stages, making the characteristic parameters into a data set, and dividing the data set into training data and test data according to a certain proportion;
setting a support vector machine model, training the support vector machine model by using the training data to obtain optimal support vector machine model parameters under constraint conditions, and constructing a classifier for deception detection based on the support vector machine model parameters;
and inputting the test data into the classifier to perform deception signal detection to obtain a detection result.
In a preferred example of the present application, it may be further configured that the step of extracting a plurality of feature parameters from the raw data of different stages includes:
in the signal processing stage, extracting a moving mean value and a moving variance of the composite SQM as characteristic parameters;
in the stage of observing data, extracting carrier-to-noise ratio moving average, moving variance and pseudo range Doppler consistency as characteristic parameters;
in the PVT resolving stage, extracting a pseudo-range positioning residual error and a Doppler velocity measurement residual error as characteristic parameters;
and in a PVT calculation result stage, extracting clock error and clock drift as characteristic parameters.
In a preferred example of the present application, the step of setting a support vector machine model, training the support vector machine model by using the training data, and obtaining optimal support vector machine model parameters under constraint conditions may further include:
the basic model of the support vector machine model is set as follows:
Figure BDA0003819650860000021
wherein, omega is a normal vector of the hyperplane, and b is a displacement term of the hyperplane;
obtaining a dual problem of the support vector machine model based on a Lagrange multiplier method, and obtaining a solving type of the support vector machine model by using Gaussian kernel function mapping:
Figure BDA0003819650860000022
wherein, k (x) i ,x j ) Representing a kernel function;
solving parameters of the support vector machine model
Figure BDA0003819650860000023
And an offset term
Figure BDA0003819650860000024
Figure BDA0003819650860000025
Parameters based on the support vector machine model
Figure BDA0003819650860000026
And an offset term
Figure BDA0003819650860000027
A classifier classification decision function for fraud detection is obtained.
In a preferred example of the present application, the classifier classification decision function may be further configured to:
Figure BDA0003819650860000028
wherein sgn (·) is a sign function, a decision function value of 1 is represented as a positive class, and a decision function value of 1 is represented as a negative class. In a preferred example of the present application, it may be further configured that the parameter for solving the support vector machine model
Figure BDA0003819650860000029
And an offset term
Figure BDA00038196508600000210
Comprises the following steps:
training a support vector machine model based on training data, selecting two parameters alpha in each circulation process by adopting a minimum sequence optimization algorithm to carry out optimization processing, and obtaining model parameters which enable an objective function to be optimal under constraint conditions through multiple circulation
Figure BDA0003819650860000031
And an offset term
Figure BDA0003819650860000032
In a preferred example of the present application, the step of generating a data set from the plurality of characteristic parameters and dividing the data set into training data and test data according to a certain proportion may further include:
subjecting the data set to data pre-processing, the pre-processing comprising a normalization operation and a principal component analysis operation,
the preprocessing operation adopts a zero-mean value standardization method, and is represented as follows:
Figure BDA0003819650860000033
in the formula,
Figure BDA0003819650860000034
represents the average of all data of the ith feature of the s-th visible satellite,
Figure BDA0003819650860000035
represents the standard deviation of all data of the ith characteristic of the s-th visible satellite,
Figure BDA0003819650860000036
which represents the initial data of the data to be transmitted,
Figure BDA0003819650860000037
representing the normalized data;
the principal component analysis operation employs the PCA method, represented as:
Figure BDA0003819650860000038
wherein, U is a principal component transfer matrix of the eigenvector, x represents a data matrix to be processed, and x = [ x ] 1 ,x 2 ,...,x n ]。
In a preferred example of the present application, the step of inputting the test data into the classifier for spoof signal detection to obtain a detection result may further include:
and inputting the test data into a classifier, wherein if the classifier outputs a first numerical value, the classifier is marked as a deception signal, and if the classifier outputs a second numerical value, the classifier is marked as a real signal.
In a second aspect, an embodiment of the present application provides a GNSS multi-parameter spoofing interference detecting apparatus, including:
the data acquisition module is used for reading acquisition signals of a GNSS deception scene and acquiring original data of different stages from the acquisition signals;
the characteristic parameter extraction module is used for extracting a plurality of characteristic parameters from the original data obtained from the data acquisition module, making the characteristic parameters into a data set, and dividing the data set into training data and test data according to a certain proportion;
the model construction module is used for setting a support vector machine model, training the support vector machine model by using the training data to obtain optimal support vector machine model parameters under constraint conditions, and constructing a classifier for deception detection based on the support vector machine model parameters;
and the detection module is used for inputting the test data into the classifier to carry out deception signal detection so as to obtain a detection result.
In a third aspect, an embodiment of the present application provides a computer device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and when the processor executes the computer program, the processor implements the steps of the GNSS spoofing interference detection method according to any one of the above items.
In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, where a computer program is stored, and the computer program, when executed by a processor, implements the steps of the GNSS spoofing interference detection method according to any one of the above-mentioned items.
In summary, compared with the prior art, the technical solution provided in the embodiments of the present application has at least the following beneficial effects: extracting a plurality of characteristic parameters from the original data of different stages in the acquired signals, and fully utilizing the advantages of satellite navigation PVT calculation; the moving mean value and the moving variance of the composite SQM are extracted from a signal processing stage, the carrier-to-noise ratio moving mean value, the moving variance and the pseudorange Doppler consistency are extracted from an observation data stage, the pseudorange positioning residual error and the Doppler velocity measurement residual error are extracted from a PVT resolving stage, the clock error and the clock drift are extracted from a PVT resolving result stage to serve as characteristic parameters, the advantages of a plurality of characteristic parameters of different stages are fused, better detection performance is obtained, the method can be well applied to more scenes, and the problem that the conventional method for detecting cheating by using single characteristic parameters is poor in applicability is solved.
Drawings
FIG. 1 is a flowchart illustrating a GNSS spoofing interference detection method according to an exemplary embodiment of the present application;
fig. 2 is a schematic flow chart of a support vector machine solution algorithm provided in another exemplary embodiment of the present application.
Detailed Description
The present embodiment is only for explaining the present application, and it is not limited to the present application, and those skilled in the art can make modifications of the present embodiment without inventive contribution as needed after reading the present specification, but all of them are protected by patent law within the scope of the claims of the present application.
In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are some embodiments of the present application, but not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
In addition, the term "and/or" in this application is only one kind of association relationship describing the association object, and means that there may be three kinds of relationships, for example, a and/or B, and may mean: a exists alone, A and B exist simultaneously, and B exists alone. In addition, the character "/" in this application generally indicates that the preceding and following related objects are in an "or" relationship, unless otherwise specified.
The terms "first," "second," and the like in this application are used for distinguishing between similar items and items that have substantially the same function or similar functionality, and it should be understood that "first," "second," and "nth" do not have any logical or temporal dependency or limitation on the number or order of execution.
The term "at least one" in this application refers to one or more, "a plurality" means two or more, for example, a plurality of first locations refers to two or more first locations.
The embodiments of the present application will be described in further detail with reference to the drawings attached hereto.
Referring to fig. 1, in an embodiment of the present application, a GNSS spoofing interference detection method is provided, the main steps of the method are described as follows:
s10: and reading the acquisition signal of the GNSS deception scene, and acquiring the original data of different stages from the acquisition signal.
And processing the read acquisition signal of the GNSS deception scene by using a software receiver, operating a signal acquisition module, a signal tracking module, a message analysis module and a PVT (virtual reality test) calculation module of the receiver, and acquiring the original data of different stages from the acquisition signal. Specifically, the raw data of the signal processing stage is obtained from the acquired signal through a signal capturing module of the receiver, the raw observation data of the observation data stage is obtained from the acquired signal through a signal tracking module, the raw data of the PVT resolving stage is obtained from the acquired signal through a text parsing module of the receiver, and the raw data of the PVT resolving result stage is obtained from the acquired signal through the PVT resolving module. PVT (Position Velocity and Time) is a solution to Position Velocity and Time information.
S20: extracting a plurality of characteristic parameters from original data in different stages, making the characteristic parameters into a data set, and dividing the data set into training data and test data according to a certain proportion.
Specifically, a plurality of characteristic parameters are extracted from original data acquired in a signal processing stage, an observation data stage, a PVT resolving stage and a PVT resolving result stage, and the characteristic parameters and a data tag which indicates whether a deception signal actually exists or not are made into a data set. From the data set, 70% was randomly selected as training data and the remaining 30% was selected as test data. Obtaining a sample set D comprising m training data and n test data, wherein
D={(x 1 ,y 1 ),(x 2 ,y 2 ),…,(x m ,y m ),…,(x m+n ,y m+n )},
Wherein x i K for the ith receiver output is the eigenvector, x i ={f 1 (i),f 2 (i),…,f k (i)},y i To indicate whether a data tag of a spoofing signal exists actually, defining a positive class tag as detecting that a receiver is affected by the spoofing signal, and defining a negative class tag as not affected by the spoofing signal, that is, only receiving a real signal, can be simplified as follows:
Figure BDA0003819650860000051
optionally, the characteristic parameter is made into a data set together with a data tag indicating whether a spoof signal is actually present. The data set is subjected to data pre-processing, including normalization operations and principal component analysis operations. Specifically, the zero-mean normalization method is adopted in this embodiment to perform normalization preprocessing, which can be expressed as:
Figure BDA0003819650860000061
wherein,
Figure BDA0003819650860000062
represents the average of all data of the ith feature of the s-th visible satellite,
Figure BDA0003819650860000063
representing the ith feature of the s-th visible satelliteThe standard deviation of all the data was found to be,
Figure BDA0003819650860000064
the initial data is represented by a representation of,
Figure BDA0003819650860000065
indicating the normalized data. In the deception classification detection, the magnitude of the dimension and the numerical value of a plurality of extracted characteristic parameters are different, and different characteristic parameters can have the same scale through standardization processing. Dimensional influence among indexes is eliminated, data processing precision is improved, and classification detection results are further accurate.
Principal Component Analysis (PCA) is adopted to carry out Principal Component Analysis, principal Component interpretation variance is obtained through singular value decomposition, the first k Principal components are selected according to the criterion that the cumulative interpretation variance ratio reaches a certain threshold, and a conversion matrix is constructed to carry out dimensionality reduction treatment:
Figure BDA0003819650860000066
wherein, U is a principal component transfer matrix of the eigenvector, x represents the data matrix to be processed, and x = [ x ] 1 ,x 2 ,...,x n ]. The dimensionality reduction is carried out on the high-dimensional data through principal component analysis, more characteristics which are convenient to understand can be found through the dimensionality reduction, the data set is easier to use, redundancy and noise of the data set are removed, and the calculated amount of an algorithm is reduced.
Optionally, in the signal processing stage, the moving mean and the moving variance of the composite SQM are extracted as characteristic parameters; in the data observation stage, extracting a carrier-to-noise ratio moving average value, a moving variance and pseudo range Doppler consistency as characteristic parameters; in the PVT resolving stage, extracting pseudo-range positioning residual errors and Doppler velocity measurement residual errors as characteristic parameters; and in a PVT calculation result stage, extracting clock error and clock drift as characteristic parameters.
Specifically, signal capture and tracking are mainly realized in the signal processing stage, and no matter synchronous interference or asynchronous interference exists, deception signals are distorted in the invasion and guide links. The Signal Quality Monitoring (SQM) algorithm is computed from the output of early, late and instantaneous correlators in the GNSS receiver tracking loop. Conventional Signal Quality Monitoring (SQM) usually selects a certain detection quantity to derive its threshold, and the statistics usually used for SQM monitoring are Ratio, delta and ELP, and the specific formula is as follows:
Figure BDA0003819650860000067
Figure BDA0003819650860000068
where the I and Q branches are output by the code phase correlators in-phase and quadrature branches, table E below indicates the early correlator, L the late correlator, P the instantaneous correlator, and d the spacing of the early correlator E and the late correlator.
In this embodiment, the composite SQM method using sliding window moving variance, the formula is as follows:
Figure BDA0003819650860000071
Figure BDA0003819650860000072
wherein k is epoch interval, w represents sliding window size, alpha is composite SQM weighting coefficient, f 1 (n) and f 2 (n) represent two statistics of the composite SQM, respectively.
Selecting a composite SQM statistic f 1 (n) and f 2 And (n) as characteristic parameters.
In the observation data stage, the observation data of the receiver mainly comprises navigation message parameters and original observation parameters. The original observation parameters comprise carrier-to-noise ratio, pseudo range, carrier phase and Doppler frequency shift information, and the carrier-to-noise ratio moving average, moving variance and pseudo range Doppler consistency are selected as characteristic parameters. The carrier-to-noise ratio C/NO reflects the signal strength of the received signal in each tracking channel, and can detect the power change during deception interference intrusion based on the carrier-to-noise ratio change to construct the moving average and moving variance of the carrier-to-noise ratio. The relationship between pseudorange and doppler consistency is constructed as follows:
f 5 (n)=ρ(n)-p(n-1)+Δt·λ·d(n)
in the formula, ρ (n) represents a pseudo-range observation value, d (n) represents a doppler observation value, and λ represents a wavelength of a signal frequency point.
In the PVT solution phase, the present embodiment selects the pseudorange positioning residual and the doppler velocity measurement residual as the characteristic parameters of the PVT solution layer. Specifically, taking pseudo-range least square positioning as an example, assuming that there are m visible satellites in total, then
Y m×1 =A m×4 ·X 4×1 +b m×1
Wherein, X is 4 multiplied by 1 dimension parameter to be estimated, Y is m multiplied by 1 dimension pseudo range measured value, A is m multiplied by 4 dimension measurement matrix, b is m multiplied by 1 dimension other error item, then least square estimated value:
Figure BDA0003819650860000073
pseudorange positioning residuals are V:
V=[I-A·(A T ·A) -1 ·A T ]·(Y-b)
the doppler velocity measurement residual is R,
R=[I-A·(A T ·A) -1 ·A T ]·(Y′-b′),
wherein, Y 'is the m × 1 dimension pseudo range change rate, b' is the m × 1 dimension velocity measurement error term,
the pseudorange positioning residual of the s-th satellite is f (n) = V S (n), the Doppler velocity measurement residual error of the s-th satellite is f (n) = R S (n)。
The most direct influence of the receiver on the deception jamming is that the PVT calculation result changes, the positioning and speed measurement result changes are related to the state of the receiver, the reference position and speed are generally required to be known for error estimation, and the time service result is generally stable in a period of time, so the deception jamming detection can be performed through the clock difference and frequency drift changes in the calculation result of the receiver, and the clock difference and the clock drift are selected as the characteristic parameters of the calculation result layer:
the clock error is as follows:
Figure BDA0003819650860000081
the clock float is as follows:
Figure BDA0003819650860000082
wherein,
Figure BDA0003819650860000083
which represents the clock difference of the receiver clock,
Figure BDA0003819650860000084
represents the clock drift of the receiver clock and c represents the speed of light.
Extracting a moving mean value and a moving variance of the composite SQM from a signal processing stage, extracting carrier-to-noise ratio moving mean value, moving variance and pseudo range Doppler consistency from an observation data stage, extracting pseudo range positioning residual error and Doppler velocity measurement residual error from a PVT resolving stage, extracting clock error and clock drift from a PVT resolving result stage as characteristic parameters, fusing the advantages of a plurality of characteristic parameters of different stages, obtaining better detection performance and improving detection accuracy; meanwhile, information characteristics of a plurality of parameters are combined, the method is not limited to the change condition of a single characteristic, can be well applied to more scenes, and solves the problem of poor scene applicability of a single-characteristic deception detection algorithm.
S30: setting a support vector machine model, training the support vector machine by using training data to obtain optimal support vector machine model parameters under constraint conditions, and constructing a classifier for deception detection based on the support vector machine model parameters.
It should be noted that the main objective of the support vector machine is to solve the separation that can accurately partition the training data and maximize the geometric separationHyperplane omega T x + b =0, wherein ω = (ω =) 1 ,ω 2 ,…,ω d ) Is the normal vector of the hyperplane, and b is the displacement term of the hyperplane. Referring to fig. 2, a schematic flow chart of a solution algorithm of a support vector machine is provided.
The basic model of the support vector machine is:
Figure BDA0003819650860000085
the dual problem of the basic model of the support vector machine is obtained by using a Lagrange multiplier method. In GNSS spoofing signal classification, the output characteristic parameter data is non-linear due to the processing of the signals through different stages of the receiver. The original space is further mapped to a higher-dimensional feature space by a gaussian kernel function, so that the training data is linearly separable in the higher-dimensional feature space. Let φ (x) be the x-mapped feature vector, the kernel function can be expressed as:
κ(x i ,x j )=<φ(x i ),φ(x j )>=φ T (x i )·φ(x j ),
wherein x is i And x j Respectively representing two different input vectors, x i Representing the ith input vector, x j Which represents the j-th input vector,<·,·>denotes the inner product operation, k (x) i ,x j ) The inner product operation of the two vectors can be replaced; specifically, the kernel function is selected as a gaussian kernel function:
Figure BDA0003819650860000091
the feature vector of the Gaussian kernel function has infinite dimensionality, the model is simple, and the obtained model is smoother.
Further replace the loss function l with a soft interval Hinge h (z) = max (0, 1-z), resulting in a solved version of the support vector machine:
Figure BDA0003819650860000093
wherein C represents a regularization constant, C is larger than 0, and alpha is a Lagrangian factor variable corresponding to the normal vector. Solving the solution type of the support vector machine by using an optimization algorithm to obtain the parameters of the support vector machine
Figure BDA0003819650860000094
And an offset term
Figure BDA0003819650860000095
Offset term
Figure BDA0003819650860000096
Can be expressed as:
Figure BDA0003819650860000097
support vector machine parameter based on solution
Figure BDA0003819650860000098
And an offset term
Figure BDA0003819650860000099
Constructing a classification decision function for a classifier for fraud detection:
Figure BDA00038196508600000910
in the two-classification problem, the decision function value is represented as a positive class when being 1, and is represented as a negative class when being 1. The use of soft intervals allows individual samples to not satisfy the constraints, thereby removing some noise or dealing with the problem that hard intervals cannot be classified, and making the data still linearly separable when mapped to high dimensions.
Optionally, the support vector machine is trained by using training data, the support vector machine is optimized by using a minimum sequence optimization algorithm SMO, and two alphas are selected in each loopPerforming optimization processing, specifically, firstly initializing alpha corresponding to all samples in the training data to be 0, traversing all samples, judging whether alpha corresponding to the current sample can be optimized, if so, randomly searching alpha corresponding to another sample, simultaneously performing optimization processing on alpha values of the two samples, and after all data are circularly traversed, finding out model parameter which enables an objective function to be optimal under a constraint condition
Figure BDA00038196508600000911
And an offset term
Figure BDA00038196508600000912
A classifier of fraud detection is constructed. By using the minimum sequence optimization algorithm, when the training sample is large, the convergence rate is still high.
S40: and inputting the test data into a classifier to perform deception signal detection to obtain a detection result.
Inputting the test data into a classifier, judging whether the test data has a deception signal or not according to the output value of the classifier, marking the output value of the classifier as the deception signal, and marking the output value of the classifier as the real signal. Specifically, a spoof signal is labeled when the classifier output value is 1, and a true signal is labeled when the classifier output value is-1. Comparing the classification result with a data label which is corresponding to the test data and indicates whether a deception signal actually exists or not, judging whether the classification result of the classifier is correct or not, counting the accuracy of classification of the classifier, wherein the accuracy reaches a preset requirement, and the classifier can be used for detecting the deception signal; if the accuracy rate does not meet the preset requirement, the support vector machine needs to be further optimized by using training data to construct a classifier.
It should be understood that, the sequence numbers of the steps in the foregoing embodiments do not imply an execution sequence, and the execution sequence of each process should be determined by functions and internal logic of the process, and should not constitute any limitation to the implementation process of the embodiments of the present application.
In one embodiment of the present application, a GNSS multi-parameter spoofing interference detecting device is provided, which corresponds one-to-one to the GNSS spoofing interference detecting method in the above embodiment. The data processing device comprises a data acquisition module, a data processing module and a data processing module, wherein the data acquisition module is used for reading acquisition signals of a GNSS deception scene and acquiring original data of different stages from the acquisition signals;
the characteristic parameter extraction module is used for extracting a plurality of characteristic parameters from the original data obtained from the data acquisition module, making the characteristic parameters into a data set, and dividing the data set into training data and test data according to a certain proportion;
the model construction module is used for constructing a support vector machine model, training the support vector machine by using training data, and obtaining the optimal support vector machine under the constraint condition as a classifier for deception detection;
and the detection module is used for inputting the test data into the classifier to carry out deception signal detection so as to obtain a detection result.
The modules of the GNSS multi-parameter spoofing interference detecting apparatus described above may be implemented in whole or in part by software, hardware, and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.
In one embodiment of the embodiments of the present application, a computer device is provided, which may be a server. The computer device includes a processor, a memory, and a network interface connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device may be implemented by any type or combination of volatile or non-volatile storage devices, including but not limited to: magnetic disk, optical disk, EEPROM (Electrically-Erasable Programmable Read Only Memory), EPROM (Erasable Programmable Read Only Memory), SRAM (Static Random Access Memory), ROM (Read-Only Memory), magnetic Memory, flash Memory, PROM (Programmable Read-Only Memory). The memory of the computer device provides an environment for the running of an operating system and computer programs stored within it. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program, when being executed by a processor, implements the steps of the GNSS spoofing interference detection method described in the above embodiments.
In an embodiment of the present application, a computer-readable storage medium is provided, which stores a computer program, which when executed by a processor, implements the GNSS spoofing interference detection method steps described in the above embodiment. The computer-readable storage medium includes a ROM (Read-Only Memory), a RAM (Random-Access Memory), a CD-ROM (Compact Disc Read-Only Memory), a magnetic disk, a floppy disk, and the like.
It will be apparent to those skilled in the art that, for convenience and brevity of description, only the above-mentioned division of each functional unit or module is illustrated, and in practical applications, the above-mentioned function may be distributed as different functional units or modules as required, that is, the internal structure of the apparatus described in this application may be divided into different functional units or modules to implement all or part of the above-mentioned functions.

Claims (10)

1. A GNSS spoofing interference detection method, the method comprising:
reading a collected signal of a GNSS deception scene, and obtaining original data of different stages from the collected signal;
extracting a plurality of characteristic parameters from the original data of different stages, making the characteristic parameters into a data set, and dividing the data set into training data and test data according to a certain proportion;
setting a support vector machine model, training the support vector machine model by using the training data to obtain optimal support vector machine model parameters under constraint conditions, and constructing a classifier for deception detection based on the support vector machine model parameters;
and inputting the test data into the classifier to perform deception signal detection to obtain a detection result.
2. The GNSS spoofing interference detecting method of claim 1, wherein the step of extracting a plurality of feature parameters from raw data of different phases comprises:
in the signal processing stage, the moving mean value and the moving variance of the composite SQM are extracted as characteristic parameters;
in the stage of observing data, extracting carrier-to-noise ratio moving average, moving variance and pseudo range Doppler consistency as characteristic parameters;
in the PVT resolving stage, extracting pseudo-range positioning residual errors and Doppler velocity measurement residual errors as characteristic parameters;
and in a PVT calculation result stage, extracting clock difference and clock drift as characteristic parameters.
3. The GNSS spoof interference detecting method of claim 1, wherein the step of setting a support vector machine model, training the support vector machine model using the training data to obtain support vector machine model parameters optimal under a constraint condition, and constructing a classifier for spoof detection based on the support vector machine model parameters comprises:
the basic model of the support vector machine model is set as follows:
Figure FDA0003819650850000011
wherein, omega is a normal vector of the hyperplane, and b is a displacement term of the hyperplane;
obtaining a dual problem of the support vector machine model based on a Lagrange multiplier method, and obtaining a solving type of the support vector machine model by using Gaussian kernel function mapping, wherein the solving type is as follows:
Figure FDA0003819650850000012
wherein, k (x) i ,x j ) Representing a kernel function;
solving parameters of the support vector machine model
Figure FDA0003819650850000021
And an offset term
Figure FDA0003819650850000022
Figure FDA0003819650850000023
Parameters based on the support vector machine model
Figure FDA0003819650850000024
And an offset term
Figure FDA0003819650850000025
A classifier classification decision function for fraud detection is obtained.
4. The GNSS spoof interference detection method of claim 3 wherein the classifier classification decision function is:
Figure FDA0003819650850000026
wherein sgn (·) is a sign function, the decision function value of-1 is represented as a positive class, and the decision function value of-1 is represented as a negative class.
5. The GNSS deception interference detection method according to claim 3, wherein the solving of the parameters of the support vector machine model
Figure FDA0003819650850000027
And an offset term
Figure FDA0003819650850000028
Comprises the following steps:
training a support vector machine model based on training data, selecting two parameters alpha in each circulation process by adopting a minimum sequence optimization algorithm to carry out optimization processing, and obtaining model parameters which enable an objective function to be optimal under constraint conditions through multiple circulations
Figure FDA0003819650850000029
And an offset term
Figure FDA00038196508500000210
6. The GNSS spoofing interference detection method of claim 1, wherein the step of forming the plurality of characteristic parameters into a data set and dividing the data set into training data and test data according to a certain ratio further comprises:
subjecting the data set to data pre-processing, the pre-processing comprising a normalization operation and a principal component analysis operation,
the preprocessing operation adopts a zero-mean value standardization method, and is represented as follows:
Figure FDA00038196508500000211
in the formula,
Figure FDA00038196508500000212
represents the average of all data of the ith feature of the s-th visible satellite,
Figure FDA00038196508500000213
represents the standard deviation, f, of all data of the ith characteristic of the s-th visible satellite i s (n) represents the initial data of the image,
Figure FDA00038196508500000214
representing the normalized data;
the principal component analysis operation employs the PCA method, represented as:
Figure FDA00038196508500000215
wherein, U is a principal component transfer matrix of the eigenvector, x represents the data matrix to be processed, and x = [ x ] 1 ,x 2 ,...,x n ]。
7. The GNSS spoof interference detecting method of claim 1, wherein the step of inputting the test data into the classifier for spoof signal detection to obtain a detection result comprises:
and inputting the test data into a classifier, and marking the classifier as a deception signal if outputting a first numerical value and marking the classifier as a real signal if outputting a second numerical value.
8. A GNSS multi-parameter spoofing interference detection apparatus, the apparatus comprising:
the data acquisition module is used for reading acquisition signals of a GNSS deception scene and acquiring original data of different stages from the acquisition signals;
the characteristic parameter extraction module is used for extracting a plurality of characteristic parameters from the original data obtained from the data acquisition module, making the characteristic parameters into a data set, and dividing the data set into training data and test data according to a certain proportion;
the model construction module is used for setting a support vector machine model, training the support vector machine model by using the training data to obtain optimal support vector machine model parameters under constraint conditions, and constructing a classifier for deception detection based on the support vector machine model parameters;
and the detection module is used for inputting the test data into the classifier to carry out deception signal detection so as to obtain a detection result.
9. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, the processor implementing the steps of the GNSS spoofing interference detection method of any one of claims 1 to 7 when executing the computer program.
10. A computer-readable storage medium, characterized in that the computer-readable storage medium stores a computer program which, when being executed by a processor, carries out the steps of the GNSS spoofing interference detection method of any one of claims 1 to 7.
CN202211038322.6A 2022-08-29 2022-08-29 GNSS deception jamming detection method and device, electronic equipment and storage medium Pending CN115494526A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202211038322.6A CN115494526A (en) 2022-08-29 2022-08-29 GNSS deception jamming detection method and device, electronic equipment and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202211038322.6A CN115494526A (en) 2022-08-29 2022-08-29 GNSS deception jamming detection method and device, electronic equipment and storage medium

Publications (1)

Publication Number Publication Date
CN115494526A true CN115494526A (en) 2022-12-20

Family

ID=84466577

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202211038322.6A Pending CN115494526A (en) 2022-08-29 2022-08-29 GNSS deception jamming detection method and device, electronic equipment and storage medium

Country Status (1)

Country Link
CN (1) CN115494526A (en)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116087994A (en) * 2023-04-07 2023-05-09 中国人民解放军战略支援部队航天工程大学 Deception jamming detection method based on machine learning
CN116482720A (en) * 2023-06-26 2023-07-25 山东科技大学 GNSS deception jamming detection method based on machine learning theory
CN116577808A (en) * 2023-07-11 2023-08-11 中国人民解放军战略支援部队航天工程大学 Navigation deception jamming detection method based on output of receiver correlator
CN116736340A (en) * 2023-04-11 2023-09-12 中山大学·深圳 Deception signal detection method, deception signal detection device, computer equipment and storage medium
CN117647826A (en) * 2024-01-29 2024-03-05 成都星历科技有限公司 Navigation deception jamming signal detection system and method based on jamming source positioning

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105550702A (en) * 2015-12-09 2016-05-04 哈尔滨工业大学 GNSS deception jamming recognition method based on SVM
CN112327331A (en) * 2020-11-02 2021-02-05 中山大学 GNSS deception jamming detection method, device, equipment and storage medium
CN113359158A (en) * 2021-06-15 2021-09-07 东南大学 GNSS generated deception jamming detection method based on SVM
CN113376659A (en) * 2021-06-15 2021-09-10 东南大学 GNSS (Global navigation satellite System) generated deception jamming detection method based on BP (Back propagation) neural network
CN113447959A (en) * 2021-06-25 2021-09-28 中山大学 GNSS deception jamming detection method based on Doppler frequency and related device

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105550702A (en) * 2015-12-09 2016-05-04 哈尔滨工业大学 GNSS deception jamming recognition method based on SVM
CN112327331A (en) * 2020-11-02 2021-02-05 中山大学 GNSS deception jamming detection method, device, equipment and storage medium
CN113359158A (en) * 2021-06-15 2021-09-07 东南大学 GNSS generated deception jamming detection method based on SVM
CN113376659A (en) * 2021-06-15 2021-09-10 东南大学 GNSS (Global navigation satellite System) generated deception jamming detection method based on BP (Back propagation) neural network
CN113447959A (en) * 2021-06-25 2021-09-28 中山大学 GNSS deception jamming detection method based on Doppler frequency and related device

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
卢丹等: "GNSS欺骗干扰检测方法", 信号处理, vol. 38, no. 6, 30 June 2022 (2022-06-30), pages 1325 - 1332 *

Cited By (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116087994A (en) * 2023-04-07 2023-05-09 中国人民解放军战略支援部队航天工程大学 Deception jamming detection method based on machine learning
CN116087994B (en) * 2023-04-07 2023-07-07 中国人民解放军战略支援部队航天工程大学 Deception jamming detection method based on machine learning
CN116736340A (en) * 2023-04-11 2023-09-12 中山大学·深圳 Deception signal detection method, deception signal detection device, computer equipment and storage medium
CN116482720A (en) * 2023-06-26 2023-07-25 山东科技大学 GNSS deception jamming detection method based on machine learning theory
CN116482720B (en) * 2023-06-26 2023-09-05 山东科技大学 GNSS deception jamming detection method based on machine learning theory
CN116577808A (en) * 2023-07-11 2023-08-11 中国人民解放军战略支援部队航天工程大学 Navigation deception jamming detection method based on output of receiver correlator
CN116577808B (en) * 2023-07-11 2023-09-29 中国人民解放军战略支援部队航天工程大学 Navigation deception jamming detection method based on output of receiver correlator
CN117647826A (en) * 2024-01-29 2024-03-05 成都星历科技有限公司 Navigation deception jamming signal detection system and method based on jamming source positioning
CN117647826B (en) * 2024-01-29 2024-04-12 成都星历科技有限公司 Navigation deception jamming signal detection system and method based on jamming source positioning

Similar Documents

Publication Publication Date Title
CN115494526A (en) GNSS deception jamming detection method and device, electronic equipment and storage medium
Wang et al. Ship detection in SAR images via local contrast of Fisher vectors
Cui et al. Statistical wavelet subband modeling for multi-temporal SAR change detection
CN113359158B (en) GNSS (Global navigation satellite System) generated deception jamming detection method based on SVM (support vector machine)
CN106295498B (en) Remote sensing image target area detection device and method
CN116492634B (en) Standing long jump testing method based on image visual positioning
CN111275416A (en) Digital currency abnormal transaction detection method and device, electronic equipment and medium
CN112596964A (en) Disk failure prediction method and device
CN113970761A (en) Non-line-of-sight signal identification method, system, computer equipment and storage medium
CN108931789A (en) Attack detection method, attack detectors, computer readable storage medium and terminal
CN112906586B (en) Time sequence action nomination generation method and related product
CN112213697B (en) Feature fusion method for radar deception jamming recognition based on Bayesian decision theory
CN111373393A (en) Image retrieval method and device and image library generation method and device
Ralph et al. An image metric-based ATR performance prediction testbed
Pelliza et al. Optimal Canny’s parameters regressions for coastal line detection in satellite-based SAR images
CN107767401B (en) Infrared target real-time tracking method and device based on nuclear correlation filtering
Chabbi et al. CFAR-CNN detector of ships from SAR image using generalized gamma distribution and real dataset
CN116184445A (en) GNSS spoofing signal detection method, device and equipment based on multiple frequency points
CN103903258A (en) Method for detecting changes of remote sensing image based on order statistic spectral clustering
CN111723686B (en) Signal number estimation method and device, electronic equipment and storage medium
JP7056751B2 (en) Ship detection systems, methods and programs
CN111950629A (en) Method, device and equipment for detecting confrontation sample
CN115201874A (en) Data quality monitoring method, device, equipment and storage medium
Jia et al. An unsupervised change detection based on automatic relationship analysis
Xu et al. Dual Attention-Based Global-Local Feature Extraction Network for Unsupervised Change Detection in PolSAR Images

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