CN116736340A - Deception signal detection method, deception signal detection device, computer equipment and storage medium - Google Patents
Deception signal detection method, deception signal detection device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN116736340A CN116736340A CN202310414982.8A CN202310414982A CN116736340A CN 116736340 A CN116736340 A CN 116736340A CN 202310414982 A CN202310414982 A CN 202310414982A CN 116736340 A CN116736340 A CN 116736340A
- Authority
- CN
- China
- Prior art keywords
- navigation
- sequence data
- data
- network
- signal detection
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 109
- 238000000034 method Methods 0.000 claims abstract description 63
- 238000013528 artificial neural network Methods 0.000 claims abstract description 61
- 238000012544 monitoring process Methods 0.000 claims abstract description 25
- 238000013527 convolutional neural network Methods 0.000 claims abstract description 24
- 230000008859 change Effects 0.000 claims abstract description 10
- 230000003042 antagnostic effect Effects 0.000 claims abstract description 4
- 238000012549 training Methods 0.000 claims description 38
- 238000012360 testing method Methods 0.000 claims description 34
- 239000011159 matrix material Substances 0.000 claims description 19
- 238000006243 chemical reaction Methods 0.000 claims description 18
- 238000004590 computer program Methods 0.000 claims description 12
- 230000007704 transition Effects 0.000 claims description 10
- 230000008485 antagonism Effects 0.000 claims description 6
- 230000000694 effects Effects 0.000 abstract description 8
- 230000006870 function Effects 0.000 description 11
- 230000008569 process Effects 0.000 description 11
- 238000012545 processing Methods 0.000 description 6
- 238000000605 extraction Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 238000011176 pooling Methods 0.000 description 4
- 238000004422 calculation algorithm Methods 0.000 description 3
- 238000010606 normalization Methods 0.000 description 3
- YLOCGHYTXIINAI-XKUOMLDTSA-N (2s)-2-amino-3-(4-hydroxyphenyl)propanoic acid;(2s)-2-aminopentanedioic acid;(2s)-2-aminopropanoic acid;(2s)-2,6-diaminohexanoic acid Chemical compound C[C@H](N)C(O)=O.NCCCC[C@H](N)C(O)=O.OC(=O)[C@@H](N)CCC(O)=O.OC(=O)[C@@H](N)CC1=CC=C(O)C=C1 YLOCGHYTXIINAI-XKUOMLDTSA-N 0.000 description 2
- 108010072051 Glatiramer Acetate Proteins 0.000 description 2
- 230000004913 activation Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 238000006073 displacement reaction Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 229940042385 glatiramer Drugs 0.000 description 2
- 238000003384 imaging method Methods 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 210000002569 neuron Anatomy 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 239000000654 additive Substances 0.000 description 1
- 230000000996 additive effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000001010 compromised effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000007637 random forest analysis Methods 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
- 238000012956 testing procedure Methods 0.000 description 1
- 238000002759 z-score normalization Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO 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
- G01S19/00—Satellite radio beacon positioning systems; Determining position, velocity or attitude using signals transmitted by such systems
- G01S19/01—Satellite radio beacon positioning systems transmitting time-stamped messages, e.g. GPS [Global Positioning System], GLONASS [Global Orbiting Navigation Satellite System] or GALILEO
- G01S19/13—Receivers
- G01S19/21—Interference related issues ; Issues related to cross-correlation, spoofing or other methods of denial of service
- G01S19/215—Interference related issues ; Issues related to cross-correlation, spoofing or other methods of denial of service issues related to spoofing
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO 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
- G01S19/00—Satellite radio beacon positioning systems; Determining position, velocity or attitude using signals transmitted by such systems
- G01S19/01—Satellite radio beacon positioning systems transmitting time-stamped messages, e.g. GPS [Global Positioning System], GLONASS [Global Orbiting Navigation Satellite System] or GALILEO
- G01S19/13—Receivers
- G01S19/35—Constructional details or hardware or software details of the signal processing chain
- G01S19/37—Hardware or software details of the signal processing chain
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- 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
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- 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
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/0475—Generative networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T11/00—2D [Two Dimensional] image generation
- G06T11/001—Texturing; Colouring; Generation of texture or colour
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Biomedical Technology (AREA)
- General Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- Computing Systems (AREA)
- Signal Processing (AREA)
- Health & Medical Sciences (AREA)
- Computer Networks & Wireless Communication (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Molecular Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Position Fixing By Use Of Radio Waves (AREA)
Abstract
The application belongs to the technical field of satellite security, and discloses a deception signal detection method, a deception signal detection device, computer equipment and a storage medium, wherein the method comprises the following steps: obtaining navigation sequence data according to the received satellite navigation signals, wherein the navigation sequence data comprises information quality monitoring mobile variance, information quality monitoring mobile average value, carrier-to-noise ratio mobile variance, carrier-to-noise ratio mobile average value, pseudo-range Doppler consistency parameters, pseudo-range residual error, receiver clock error and receiver clock error change rate; converting the navigation sequence data into navigation image data; inputting the navigation image data into a trained signal detection neural network to obtain a deception signal detection result; the signal detection neural network is an antagonistic convolutional neural network. The application can achieve the effect of improving the detection accuracy of the deception signal.
Description
Technical Field
The present application relates to the field of satellite security technologies, and in particular, to a method and apparatus for detecting a spoofing signal, a computer device, and a storage medium.
Background
The development of the global navigation satellite system has been widely used in various industries, supporting many modern systems. However, satellite signals still have a number of significant drawbacks, such as being susceptible to interference and fraud. Due to the characteristic of satellite signals, the research emphasis direction of the global navigation satellite system gradually transits from the initial direction of improving the positioning precision to the direction of expanding the system application and improving the safety and reliability performance of the system, and the capability of enhancing the anti-deception jamming capability of the global navigation satellite system has become a hot spot for research in the industry and academia.
Among all interference categories, spoofing interference is the most compromised type of interference. Spoofing refers to the fact that a spoofing machine emits a spoofing signal to induce a user receiver to generate incorrect position, speed or time information, which affects the normal operation and use of the receiver and enables control of the target receiver, with serious consequences if the system uses such incorrect information.
In order to cope with the influence of the spoofing, and ensure that the global navigation satellite system provides correct navigation, positioning and timing services for the end user, many detection methods have been proposed in the art successively, for example, the C/N0 detection method finds the presence of the spoofing signal by detecting the abnormal change of C/N0, but when the spoofing signal is transmitted together with noise, misjudgment is easily caused. Therefore, the method for detecting the spoofing signal in the satellite navigation signal in the existing navigation system has the problem of poor accuracy.
Disclosure of Invention
The application provides a deception signal detection method, a deception signal detection device, a computer device and a storage medium, which can improve the accuracy of deception signal detection.
In a first aspect, an embodiment of the present application provides a fraud signal detection method, including: obtaining navigation sequence data according to the received satellite navigation signals, wherein the navigation sequence data comprises information quality monitoring mobile variance, information quality monitoring mobile average value, carrier-to-noise ratio mobile variance, carrier-to-noise ratio mobile average value, pseudo-range Doppler consistency parameters, pseudo-range residual error, receiver clock error and receiver clock error change rate; converting the navigation sequence data into navigation image data; inputting the navigation image data into a trained signal detection neural network to obtain a deception signal detection result; the signal detection neural network is an antagonistic convolutional neural network.
Further, the obtaining navigation sequence data according to the received satellite navigation signal includes: and according to the received satellite navigation signals and the resolving receiver parameters, resolving the satellite navigation signals to obtain navigation sequence data.
Further, the converting the navigation sequence data into navigation image data includes: the navigation sequence data is converted into navigation image data by adopting a Markov transition field method.
Further, the converting the navigation sequence data into navigation image data includes: the navigation sequence data is converted into navigation image data by using a method of a glamer angle field.
Further, the method for converting navigation sequence data into navigation image data by using the glatiramer angle field comprises the following steps:
normalizing the navigation sequence data to obtain scaled navigation sequence data; converting the scaled navigation sequence data from a rectangular coordinate system to a polar coordinate system to obtain polar coordinate navigation sequence data and characteristic quantity of the navigation sequence data; performing angle difference on the polar coordinate navigation sequence data to obtain a gram matrix; and generating navigation image data according to the feature quantity of the navigation sequence data and the gram matrix.
Further, the method further comprises:
constructing an anti-convolution neural network, wherein the anti-convolution neural network comprises a generation network and an identification network;
training sample acquisition: acquiring a real navigation image sample, and inputting random noise to a generation network so that the generation network generates a false sample;
training sample identification: inputting the false sample and the real navigation image sample into an identification network for identification to obtain an identification result and feedback gradient information, and updating the generation network according to the feedback gradient information to obtain an updated generation network;
repeatedly executing a training sample acquisition step and a training sample identification step based on the updated generation network until the antagonism convolutional neural network meets a first preset condition, wherein the first preset condition is that the probability that the identification network identifies the real navigation image sample as true is the same as the probability that the identification network identifies the false sample as true;
and taking the antagonism convolution neural network meeting the first preset condition as a trained signal detection neural network.
Further, the acquiring the real navigation image sample includes: obtaining training navigation sequence data based on the historical satellite navigation signals; and converting the training navigation sequence data into a real navigation image sample.
Further, the generation network and the authentication network both adopt deep convolutional neural networks.
Further, the method further comprises: acquiring a test data set, wherein the test data set comprises a first navigation image test data set obtained by a deception signal and a second navigation image test data set obtained by a real satellite navigation signal; testing the trained signal detection neural network based on the test data set to obtain a test result, wherein the test result is the detection accuracy; and when the test result does not meet the second preset condition, retraining the trained signal detection neural network.
In a second aspect, an embodiment of the present application provides a fraud signal detection apparatus, including:
the data acquisition module is used for acquiring navigation sequence data according to the received satellite navigation signals; the navigation sequence data includes: information quality monitoring mobile variance, information quality monitoring mobile mean, carrier-to-noise ratio mobile variance, carrier-to-noise ratio mobile mean, pseudo-range Doppler consistency parameter, pseudo-range residual error, receiver clock error and receiver clock error change rate;
the data conversion module is used for converting the navigation sequence data into navigation image data;
and the data detection module is used for inputting the navigation image data into the trained signal detection neural network to obtain a deception signal detection result.
Further, the data conversion module is used for converting the navigation sequence data into the navigation image data by adopting a Markov transition field method.
Further, the data conversion module is used for converting the navigation sequence data into the navigation image data by using a method of a gram angle field.
Further, the data conversion module includes:
the scaling unit is used for normalizing the navigation sequence data to obtain scaled navigation sequence data;
the coordinate system conversion unit is used for converting the scaled navigation sequence data from a rectangular coordinate system to a polar coordinate system to obtain polar coordinate navigation sequence data and characteristic quantity of the navigation sequence data;
the matrix generation unit is used for performing angle difference on the polar coordinate navigation sequence data to obtain a gram matrix;
and the image data generating unit is used for generating navigation image data according to the feature quantity of the navigation sequence data and the gram matrix.
In a third aspect, an embodiment of the present application provides a computer device including a memory, a processor, and a computer program stored on the memory and executable on the processor, the processor executing the steps of the fraud signal detection method of any of the embodiments described above when the computer program is executed.
In a fourth aspect, embodiments of the present application provide a computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps of the fraud signal detection method of any of the embodiments described above.
In summary, compared with the prior art, the technical scheme provided by the embodiment of the application has the following beneficial effects:
according to the deception signal detection method provided by the embodiment of the application, various parameters such as information quality monitoring mobile variance, information quality monitoring mobile average value, carrier-to-noise ratio mobile variance and the like in navigation sequence data are obtained, the navigation sequence data comprising the parameters are converted into navigation image data, and a trained signal detection neural network is utilized to detect the navigation image data comprising the parameters; the detection of various parameters in the navigation image data enables the detection to be more comprehensive, and the accuracy of the obtained detection result is higher; meanwhile, the operation of converting the navigation sequence data into the navigation image data fully utilizes the characteristic of the neural network that the characteristic extraction effect of the image data is better, so that the fineness and the accuracy of the detection of the signal detection neural network are better.
Drawings
Fig. 1 is a flowchart of a fraud signal detection method according to an exemplary embodiment of the present application.
Fig. 2 is a flowchart of a navigation image data acquisition step provided in an exemplary embodiment of the present application.
FIG. 3 is a flowchart of an combat convolutional neural network training step provided in accordance with an exemplary embodiment of the present application.
Fig. 4 is a flowchart of a signal detection neural network testing procedure according to an exemplary embodiment of the present application.
Fig. 5 is a block diagram of a spoofing signal detecting apparatus according to an exemplary embodiment of the present application.
Fig. 6 is a block diagram of a data conversion module according to an exemplary embodiment of the present application.
Detailed Description
The following description of the embodiments of the present application will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present application, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
Referring to fig. 1, an embodiment of the present application provides a method for detecting a spoofing signal, which is described by taking a terminal as an execution body as an example, and the method may include:
step S1, navigation sequence data is obtained according to received satellite navigation signals, wherein the navigation sequence data comprises information quality monitoring mobile variance, information quality monitoring mobile average value, carrier-to-noise ratio mobile variance, carrier-to-noise ratio mobile average value, pseudo-range Doppler consistency parameters, pseudo-range residual error, receiver clock error and receiver clock error change rate;
the satellite navigation signal may be expressed as:
S R (t)＝S T (t)+S S (t)+n 0 (t)
wherein S is R (t) represents an intermediate frequency signal received by the receiver S T (t) and S S (t) representing the true satellite signal and the rogue signal, respectively, n 0 (t) represents a mean value of 0 and a variance of sigma 2 Additive white gaussian noise of (c).
Since the spoofing signal has a similar signal structure to the true satellite signal, the true satellite signal and the spoofing signal can be expressed as follows:
wherein M and N represent the number of true satellite signals and spoofing signals in the received signals; p (P) i T And P i S The power of the ith real signal and the power of the spoofing signal are respectively represented; c (C) i (t) pseudo code representing an ith satellite; d (D) i (t) represents an ith signal navigation message data bit; f (f) IF Representing the intermediate frequency of the signal;and->The Doppler frequency shifts of the ith real satellite signal and the spoofing signal are respectively represented; />Representing the i-th signal code phase; />And->Representing the initial carrier phases of the true satellite signal and the spoofing signal, respectively.
The navigation sequence data may include various parameters such as information quality monitoring mobile variance, information quality monitoring mobile average, carrier-to-noise ratio mobile variance, carrier-to-noise ratio mobile average, pseudo-range doppler consistency parameter, pseudo-range residual, receiver clock error and receiver clock error change rate.
Step S2, the navigation sequence data are converted into navigation image data.
The navigation image data is two-dimensional navigation image data without depth information, and the two-dimensional navigation image data comprises various parameters such as information quality monitoring mobile variance, information quality monitoring mobile mean, carrier-to-noise ratio mobile variance, carrier-to-noise ratio mobile mean, pseudo-range Doppler consistency parameters, pseudo-range residual errors, receiver clock error change rate and the like in navigation sequence data.
In particular, the conversion of navigation sequence data into navigation image data may be achieved using a time-series imaging algorithm. Time-series imaging algorithms include methods of the Glatiramer Angle Field (GAF), markov Transition Field (MTF), or recursive graph (RP).
The method of the gram angle field is roughly that navigation sequence data are normalized to obtain zoom navigation sequence data; then converting the scaled navigation sequence data from a rectangular coordinate system to a polar coordinate system to obtain polar coordinate navigation sequence data and characteristic quantity of the navigation sequence data; performing angle difference on the polar coordinate navigation sequence data to obtain a gram matrix; and finally generating navigation image data according to the feature quantity of the navigation sequence data and the gram matrix.
The Markov transition field is used for identifying quantile units in navigation sequence data, and then the paired transition probabilities of the quantile units are encoded into a Markov transition matrix representing an image to obtain navigation image data.
The recursion diagram mainly proposes a framework for converting navigation sequence data into navigation image data using the recursion diagram.
Step S3, inputting the navigation image data into a trained signal detection neural network to obtain a deception signal detection result; the signal detection neural network is an antagonistic convolutional neural network.
The anti-convolution neural network generally comprises a generating network and an identifying network, the generating network is used for generating false samples according to random noise, the identifying network can identify the true navigation image samples and the false samples to obtain identification results and feedback gradient information, then the feedback gradient information is sent to the generating network, the generating network can modify network parameters in the generating network according to the feedback gradient information, the identifying network can gradually improve the self-identification performance in the identification process, and finally the false samples generated by the generating network can be deceived into the identifying network, so that the probability that the identifying network identifies the true navigation image samples as true is equal to the probability that the false samples are identified as true, and the anti-convolution neural network is a trained signal detection neural network.
Specifically, the detection result of the spoofing signal may be 0 or 1, and when the detection result of the spoofing signal is 0, it indicates that the spoofing signal does not exist in the satellite navigation signal; when the detection result of the spoofing signal is 1, the spoofing signal exists in the satellite navigation signal.
According to the deception signal detection method provided by the embodiment, various parameters such as information quality monitoring mobile variance, information quality monitoring mobile average value, carrier-to-noise ratio mobile variance and the like in navigation sequence data are obtained, the navigation sequence data comprising the parameters are converted into navigation image data, and a trained signal detection neural network is utilized to detect the navigation image data comprising the parameters; the detection of various parameters in the navigation image data enables the detection to be more comprehensive, and the accuracy of the obtained detection result is higher; meanwhile, the operation of converting the navigation sequence data into the navigation image data fully utilizes the characteristic of the neural network that the characteristic extraction effect of the image data is better, so that the fineness and the accuracy of the detection of the signal detection neural network are better.
Based on the above embodiments, in some embodiments, in step S1, obtaining navigation sequence data according to the received satellite navigation signal may include: and according to the received satellite navigation signals and the resolving receiver parameters, resolving the satellite navigation signals to obtain navigation sequence data.
The resolving receiver can be a GNSS receiver, is a common instrument in engineering detection, and has the main function of monitoring the horizontal and vertical displacement of various structures. One instrument can monitor displacement in two directions simultaneously, and the root cause is that the GNSS positioning technology is adopted.
Specifically, the GNSS receiver mainly comprises a receiver antenna unit, a host unit and a power supply, wherein the main function of the receiving antenna is to receive radio frequency signals and convert electromagnetic waves broadcast by satellites into electrical signals which are convenient to process; the main function of the host unit is to track, process and measure the processed electric signals, the geodetic GNSS receiver is mainly used for carrying out relative positioning by using carrier phase values, the signals transmitted by satellites are received firstly during working, then the pseudo range and Doppler frequency shift are calculated through internal signal processing, finally the satellite navigation message is regulated, and the three-dimensional coordinates of a user are obtained, so that the navigation positioning function is realized.
According to the embodiment, the resolving receiver in the prior art is used for resolving various parameters of the received satellite navigation signals, so that navigation sequence data of the satellite navigation signals can be obtained quickly and conveniently.
In some embodiments, step S2 may include: the navigation sequence data is converted into navigation image data by adopting a Markov transition field method.
In other embodiments, step S2 may include: the navigation sequence data is converted into navigation image data by using a method of a glamer angle field.
Referring to fig. 2, in some embodiments, step S2 may specifically further include the following steps:
step S21, normalizing the navigation sequence data to obtain zoom navigation sequence data.
The normalized formula is specifically:
wherein x is i For scaling the navigation sequence data, X is the navigation sequence data.
And S22, converting the scaled navigation sequence data from a rectangular coordinate system to a polar coordinate system to obtain polar coordinate navigation sequence data and characteristic quantity of the navigation sequence data.
The formula for converting to the polar coordinate system is as follows:
in the formula, N is the characteristic quantity of navigation sequence data, r i And phi i Is the obtained polar navigation sequence data.
And S23, performing angle difference on the polar coordinate navigation sequence data to obtain a gram matrix.
The formula for making the angle difference in the above steps is:
the form of the resulting gram matrix can be expressed as:
and step S24, generating navigation image data according to the feature quantity of the navigation sequence data and the gram matrix.
According to the embodiment, the navigation sequence data can be converted into the navigation image data through the gram angle field method, the extraction capability of the signal detection neural network to the image characteristics can be fully utilized, the detection of the deception signals in the satellite navigation signals by the signal detection neural network is enabled to be more accurate, and the obtained deception signal detection result is more accurate.
Referring to fig. 3, in some embodiments, the method may further comprise the steps of:
step S41, constructing an anti-convolution neural network, wherein the anti-convolution neural network comprises a generation network and an identification network.
Step S42, acquiring a real navigation image sample, and inputting random noise to the generation network so that the generation network generates false samples.
And step S43, inputting the false sample and the real navigation image sample into an authentication network for authentication to obtain an authentication result and feedback gradient information, updating the generation network according to the feedback gradient information, and obtaining an updated generation network.
And S44, repeatedly executing S42 and S43 based on the updated generation network until the antagonism convolutional neural network meets a first preset condition, wherein the first preset condition is that the probability that the authentication network authenticates the real navigation image sample as true is the same as the probability that the authentication network authenticates the false sample as true.
And step S45, taking the antagonism convolution neural network meeting the first preset condition as a trained signal detection neural network.
The convolutional neural network is a DCGAN network, and one main difference between the convolutional neural network and the traditional network model is that the convolutional neural network is composed of two network modules, namely a generating network and an identifying network.
The generation network and the authentication network can be said to be completely independent two models, and can be understood to be completely independent two neural networks. Thus, the training mode is generally training through single alternate iterative training. During the training process, the goal of generating the network is to generate as much real data as possible to spoof the authentication network. The goal of identifying the network is to separate the false sample generated by the generating network from the real navigation image sample as much as possible.
The value function of DCGAN is as follows:
wherein D is ω (x) For true navigation image sample, D ω (G θ (z)) is a false sample;is the probability that the real navigation image sample is authenticated as true by the authentication network,/for example>Is the probability that a false sample is identified as false by the authentication network. The purpose of the authentication network is to maximize the value function, i.e. in the above formula Maximum is required.
WhileAt the maximum, the probability that the authentication network is required to authenticate the real navigation image sample as true is close to 1, and +.>At maximum, the probability that the authentication network will be true for a false sample is required to be 0.
The purpose of generating a network is opposite to the purpose of identifying a network, and the aim is to minimize a value function and reflect the value function into a formulaAnd->The probability that the true navigation image sample is identified as true is small, and the probability that the identification network judges that the false sample is true is close to 1. If the generating network is able to achieve this goal, the false samples it generates can impersonate the authentication network with spurious.
The process is a countermeasure process, the generating network and the identifying network want to optimize the value function in a self mode, when the countermeasure balance point is 1/2 of the probability of true and false sample data, namely the probability that the identifying network identifies the true navigation image sample as true is the same as the probability that the identifying network identifies the false sample as true, balance is achieved at the moment, the training of the DCGAN network converges, and the DCGAN network at the moment is used as a trained signal detection neural network.
In the prior art, although a machine learning method such as an SVM (Support Vector Machine, a vector machine), a CNN (Convolutional Neural Networks, a convolutional neural network), a random forest and the like can be improved to a certain extent in effect compared with a traditional detection algorithm, in general, in order to obtain a better model, training of a large number of data sets is required, and in addition, the diversity of the data sets needs to be ensured, so that the model is suitable for various scenes, otherwise, the trained model may be only suitable for current data, and once the trained model is applied to a new environment, the detection method is not suitable.
The method for obtaining the trained signal detection neural network by training the DCGAN network is convenient and easy to implement, can directly realize effective detection of the deceptive signal after storing the trained signal detection neural network data, has relatively low complexity, has low requirements on a receiver, can reduce equipment cost of the receiver, and has wider application scene and higher applicability.
Meanwhile, the problem of insufficient training data sets of the traditional machine learning method is solved, the generation network of the DCGAN can generate false samples for many times in the countermeasure training process, the problem of lack of the data sets of the traditional training method is overcome to a certain extent, the labor cost for collecting a large amount of deception signal data is saved, the method is more suitable for detecting deception interference signals sent by general deception sources, and the applicability of the method is further improved.
In some embodiments, step S42 may specifically further include: obtaining training navigation sequence data based on the historical satellite navigation signals; and converting the training navigation sequence data into a real navigation image sample.
In the specific implementation process, the obtained training navigation sequence data can be converted into training navigation image data, then the offset value, the weight value, the mean value and the standard deviation of the training navigation image data are calculated, and the training navigation image data are standardized according to the offset value, the weight value, the mean value and the standard deviation of the training navigation image data, so that a real navigation image sample is obtained.
Specifically, the normalization method includes:
the normalization method is to linearly transform the training navigation image data to map the real navigation image sample to the [0,1] interval.
The normalization method is to normalize the data based on the mean (mean) and standard deviation (standard deviation) of the training navigation image data. The original value x of a is normalized to x' using z-score. The z-score normalization method is applicable to the case where the maximum value and the minimum value of the attribute a are unknown, or the case where there is outlier data out of the range of values.
According to the embodiment, the training speed of the anti-convolution neural network and the accuracy of identifying the real navigation image samples are improved by standardizing the training navigation image data.
In some embodiments, the generation network and the authentication network both employ deep convolutional neural networks.
The deep convolutional neural network mainly comprises an input layer, a convolutional layer, an activation function, a pooling layer, a full-connection layer and an output layer.
Input layer: the deep convolution network can directly take the picture as the input of the network, and extract the characteristics through training.
Convolution layer: by convolution operation it is essentially another representation of the input, and if the convolution layer is considered as a black box, the output can be considered as another representation of the input, and training of the whole network is training out the intermediate parameters required for such a representation.
The deep convolutional network connects small neural networks in series to form the deep neural network, and two special processing modes are mainly available:
local receptive fields are used: the neurons are connected only with the neurons of the next upper layer, and the last global feature is formed by combining the learned local features.
Weight sharing is adopted: when the same convolution kernel operates different local receptive fields, the same weight parameter is adopted, so that the parameter calculation amount required in the network operation process can be reduced. The different features of the picture are obtained through the convolution kernels of each layer, and specific positions of the features in the picture do not need to be considered specially, so that the processing mode has obvious advantages in the task of analyzing and processing the picture.
Pooling layer: the method is a special processing operation for the data in the convolutional neural network, and the characteristic size of the picture is reduced through pooling processing, so that the problem of large calculation amount caused by taking the result of the upper layer as input can be effectively solved.
Activation function: both convolution and pooling operations in a network are linear operations, and the large number of samples in life are not linearly related when classified, so that nonlinear elements need to be introduced into the network so that the network can solve the problem of nonlinearity.
Full tie layer: the layer is the layer with the most consumption parameters in the network, if the input of the full-connection layer is 4×4×100 and the output of the full-connection layer is 512, the layer needs 4×4×100×512 parameters; whereas a typical convolutional layer requires only 4 x 512 parameters if the convolutional kernel is 4*4 and the output is 512. A typical network will have two fully connected layers, the output of the second fully connected layer corresponding to the output of the classified number.
According to the embodiment, the generation network and the identification network replace a fully-connected network in the traditional countermeasure convolutional neural network by the deep convolutional neural network, so that the stability and the robustness of the signal detection neural network are improved, meanwhile, the extraction effect of the deep convolutional neural network on the image characteristics is better than that of the fully-connected network, the method is more suitable for detecting navigation image data, and the accuracy of the signal detection neural network on deception signal detection is effectively improved.
Referring to fig. 4, in some embodiments, the method may further comprise the steps of:
step S51, a test data set is acquired, the test data set including a first navigation image test data set obtained from the spoofing signal and a second navigation image test data set obtained from the real satellite navigation signal.
And step S52, testing the trained signal detection neural network based on the test data set to obtain a test result, wherein the test result is the detection accuracy.
And step S53, when the test result does not meet the second preset condition, retraining the trained signal detection neural network.
The test data set may also be a standardized navigation image test data set.
The second preset condition may specifically be that the accuracy is 80%, and when the test result shows that the accuracy of the current signal detection neural network on detecting the spoofing signal is less than 80%, the real navigation image sample and the random noise are input again for training.
According to the embodiment, the trained signal detection neural network is tested by using the test data set, and the detection result of the test data set by the signal detection neural network is compared with the test data set, so that the detection accuracy of the signal detection neural network can be roughly calculated; furthermore, when a plurality of trained signal detection neural networks exist, the signal detection neural network with better performance can be selected according to the detection accuracy in the test results corresponding to the signal detection neural networks, and the detection accuracy of the signal detection neural network in actual use is effectively improved.
Referring to fig. 5, another embodiment of the present application provides a fraud signal detection apparatus, which may include:
the data acquisition module 101 is configured to obtain navigation sequence data according to a received satellite navigation signal; the navigation sequence data includes: information quality monitoring mobile variance, information quality monitoring mobile mean, carrier-to-noise ratio mobile variance, carrier-to-noise ratio mobile mean, pseudorange Doppler consistency parameter, pseudorange residual, receiver clock difference and receiver clock difference change rate.
The data conversion module 102 is configured to convert the navigation sequence data into navigation image data.
The data detection module 103 is configured to input the navigation image data into a trained signal detection neural network, so as to obtain a spoofed signal detection result.
In the above-mentioned embodiment, the data acquisition module 101 acquires a plurality of parameters such as information quality monitoring mobile variance, information quality monitoring mobile average value, carrier-to-noise ratio mobile variance in the navigation sequence data, the data conversion module 102 converts the navigation sequence data including the parameters into navigation image data, and the data detection module 103 detects the navigation image data including the parameters by using the trained signal detection neural network; the detection of various parameters in the navigation image data enables the detection to be more comprehensive, and the accuracy of the obtained detection result is higher; meanwhile, the operation of converting the navigation sequence data into the navigation image data fully utilizes the characteristic of the neural network that the characteristic extraction effect of the image data is better, so that the fineness and the accuracy of the detection of the signal detection neural network are better.
In some embodiments, the data conversion module 102 may be configured to convert the navigation sequence data into the navigation image data using a markov transition field method.
In some embodiments, the data conversion module 102 may be configured to convert the navigation sequence data into the navigation image data using a method of using a glamer angle field.
Referring to fig. 6, in some embodiments, the data conversion module 102 may include:
the scaling unit 21 is configured to normalize the navigation sequence data to obtain scaled navigation sequence data.
The coordinate system converting unit 22 is configured to convert the scaled navigation sequence data from a rectangular coordinate system to a polar coordinate system, and obtain polar coordinate navigation sequence data and feature numbers of the navigation sequence data.
The matrix generating unit 23 is configured to perform angle difference on the polar navigation sequence data to obtain a gram matrix.
An image data generating unit 24 for generating navigation image data based on the feature quantity of the navigation sequence data and the gram matrix.
The above embodiment specifically provides how the data conversion module 102 converts the navigation sequence data into the navigation image data, and fully utilizes the capability of the signal detection neural network for extracting the image features, so that the detection of the spoofing signal in the satellite navigation signal by the signal detection neural network is more accurate, and the obtained spoofing signal detection result is more accurate.
The specific limitation of the apparatus for detecting a spoofing signal provided in this embodiment may be referred to the above embodiments of the method for detecting a spoofing signal, which is not described herein. The respective modules in the above-described spoofing signal detecting apparatus may be realized in whole or in part by software, hardware, and a combination thereof. The above modules may be embedded in hardware or may be independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above modules.
Embodiments of the present application provide a computer device that may include a processor, memory, network interface, and database 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 includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program, when executed by a processor, causes the processor to perform the steps of the fraud signal detection method of any of the embodiments described above.
The working process, working details and technical effects of the computer device provided in this embodiment may be referred to the above embodiments of the method for detecting a spoofing signal, which are not described herein.
An embodiment of the present application provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps of the fraud signal detection method of any of the embodiments described above. The computer readable storage medium refers to a carrier for storing data, and may include, but is not limited to, a floppy disk, an optical disk, a hard disk, a flash Memory, and/or a Memory Stick (Memory Stick), etc., where the computer may be a general purpose computer, a special purpose computer, a computer network, or other programmable devices.
The working process, working details and technical effects of the computer readable storage medium provided in this embodiment can be referred to the above embodiments of the spoofing signal detecting method, and are not repeated here.
Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in embodiments provided herein may include non-volatile and/or volatile memory. The nonvolatile memory can include Read Only Memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double Data Rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous Link DRAM (SLDRAM), memory bus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), among others.
The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.
The above examples illustrate only a few embodiments of the application, which are described in detail and are not to be construed as limiting the scope of the application. It should be noted that it will be apparent to those skilled in the art that several variations and modifications can be made without departing from the spirit of the application, which are all within the scope of the application. Accordingly, the scope of protection of the present application is to be determined by the appended claims.
Claims (15)
1. A spoofing signal detecting method, the method comprising:
obtaining navigation sequence data according to a received satellite navigation signal, wherein the navigation sequence data comprises information quality monitoring mobile variance, information quality monitoring mobile average value, carrier-to-noise ratio mobile variance, carrier-to-noise ratio mobile average value, pseudo-range Doppler consistency parameter, pseudo-range residual error, receiver clock error and receiver clock error change rate;
converting the navigation sequence data into navigation image data;
inputting the navigation image data into a trained signal detection neural network to obtain a deception signal detection result; the signal detection neural network is an antagonistic convolutional neural network.
2. The method of claim 1, wherein obtaining navigation sequence data from the received satellite navigation signals comprises: and according to the received satellite navigation signals and the received decoding receiver parameters, decoding the satellite navigation signals to obtain the navigation sequence data.
3. The method of claim 1, wherein said converting said navigation sequence data into navigation image data comprises: the navigation sequence data is converted into navigation image data by adopting a Markov transition field method.
4. The method of claim 1, wherein said converting said navigation sequence data into navigation image data comprises: the navigation sequence data is converted into navigation image data by using a method of a gram angle field.
5. The method of claim 4, wherein the method of using a glamer angle field to convert the navigation sequence data into navigation image data comprises:
normalizing the navigation sequence data to obtain scaled navigation sequence data;
converting the scaled navigation sequence data from a rectangular coordinate system to a polar coordinate system to obtain polar coordinate navigation sequence data and characteristic quantity of the navigation sequence data;
performing angle difference on the polar coordinate navigation sequence data to obtain a gram matrix;
and generating the navigation image data according to the characteristic quantity of the navigation sequence data and the gram matrix.
6. The method according to claim 1, wherein the method further comprises:
constructing an countermeasure convolutional neural network, wherein the countermeasure convolutional neural network comprises a generation network and an identification network;
training sample acquisition: acquiring a real navigation image sample, and inputting random noise to the generation network so as to enable the generation network to generate false samples;
training sample identification: inputting the false sample and the real navigation image sample into the identification network for identification to obtain an identification result and feedback gradient information, and updating the generation network according to the feedback gradient information to obtain the updated generation network;
repeating the training sample acquiring step and the training sample identifying step based on the updated generation network until the antagonism convolutional neural network meets a first preset condition, wherein the first preset condition is that the probability that the identification network identifies the real navigation image sample as true is the same as the probability that the identification network identifies the false sample as true;
and taking the antagonism convolution neural network meeting the first preset condition as the trained signal detection neural network.
7. The method of claim 6, wherein the acquiring a real navigation image sample comprises:
obtaining training navigation sequence data based on the historical satellite navigation signals;
and converting the training navigation sequence data into the real navigation image sample.
8. The method of claim 6, wherein the generating network and the discriminating network each employ a deep convolutional neural network.
9. The method according to claim 1, wherein the method further comprises:
acquiring a test data set, wherein the test data set comprises a first navigation image test data set obtained by a deception signal and a second navigation image test data set obtained by a real satellite navigation signal;
testing the trained signal detection neural network based on the test data set to obtain a test result, wherein the test result is the detection accuracy;
and when the test result does not meet a second preset condition, retraining the trained signal detection neural network.
10. A spoofing signal detecting apparatus, the apparatus comprising:
the data acquisition module is used for acquiring navigation sequence data according to the received satellite navigation signals; the navigation sequence data includes: information quality monitoring mobile variance, information quality monitoring mobile mean, carrier-to-noise ratio mobile variance, carrier-to-noise ratio mobile mean, pseudo-range Doppler consistency parameter, pseudo-range residual error, receiver clock error and receiver clock error change rate;
the data conversion module is used for converting the navigation sequence data into navigation image data;
and the data detection module is used for inputting the navigation image data into a trained signal detection neural network to obtain a deception signal detection result.
11. The apparatus of claim 10, wherein the data conversion module is configured to convert the navigation sequence data into the navigation image data using a markov transition field method.
12. The apparatus of claim 10, wherein the data conversion module is configured to convert the navigation sequence data into the navigation image data using a method of a glamer angle field.
13. The apparatus of claim 12, wherein the data conversion module comprises:
the scaling unit is used for normalizing the navigation sequence data to obtain scaled navigation sequence data;
the coordinate system conversion unit is used for converting the scaled navigation sequence data from a rectangular coordinate system to a polar coordinate system to obtain polar coordinate navigation sequence data and characteristic quantity of the navigation sequence data;
the matrix generation unit is used for carrying out angle difference on the polar coordinate navigation sequence data to obtain a gram matrix;
and the image data generating unit is used for generating the navigation image data according to the characteristic quantity of the navigation sequence data and the gram matrix.
14. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the steps of the method according to any one of claims 1 to 9 when the computer program is executed.
15. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method according to any one of claims 1 to 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310414982.8A CN116736340A (en) | 2023-04-11 | 2023-04-11 | Deception signal detection method, deception signal detection device, computer equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310414982.8A CN116736340A (en) | 2023-04-11 | 2023-04-11 | Deception signal detection method, deception signal detection device, computer equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116736340A true CN116736340A (en) | 2023-09-12 |
Family
ID=87901860
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310414982.8A Pending CN116736340A (en) | 2023-04-11 | 2023-04-11 | Deception signal detection method, deception signal detection device, computer equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116736340A (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2008024534A2 (en) * | 2006-05-16 | 2008-02-28 | Southwest Research Institute | Detection of deception signal used to deceive geolocation receiver of a satellite navigation system |
| CN108805188A (en) * | 2018-05-29 | 2018-11-13 | 徐州工程学院 | A kind of feature based recalibration generates the image classification method of confrontation network |
| CN110046653A (en) * | 2019-03-22 | 2019-07-23 | 赣州好朋友科技有限公司 | A kind of white tungsten method for separating and system based on XRT ray |
| CN111401261A (en) * | 2020-03-18 | 2020-07-10 | 金陵科技学院 | Robot gesture recognition method based on GAN-CNN framework |
| CN112327331A (en) * | 2020-11-02 | 2021-02-05 | 中山大学 | GNSS deception jamming detection method, device, equipment and storage medium |
| CN114092844A (en) * | 2021-11-19 | 2022-02-25 | 中国直升机设计研究所 | Multi-band image target detection method based on generation countermeasure network |
| CN115494526A (en) * | 2022-08-29 | 2022-12-20 | 中山大学 | GNSS deception jamming detection method and device, electronic equipment and storage medium |
| CN115598669A (en) * | 2022-09-26 | 2023-01-13 | 中山大学（Cn） | Navigation multi-feature GNSS deception jamming detection method, system, equipment and medium |
-
2023
- 2023-04-11 CN CN202310414982.8A patent/CN116736340A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2008024534A2 (en) * | 2006-05-16 | 2008-02-28 | Southwest Research Institute | Detection of deception signal used to deceive geolocation receiver of a satellite navigation system |
| CN108805188A (en) * | 2018-05-29 | 2018-11-13 | 徐州工程学院 | A kind of feature based recalibration generates the image classification method of confrontation network |
| CN110046653A (en) * | 2019-03-22 | 2019-07-23 | 赣州好朋友科技有限公司 | A kind of white tungsten method for separating and system based on XRT ray |
| CN111401261A (en) * | 2020-03-18 | 2020-07-10 | 金陵科技学院 | Robot gesture recognition method based on GAN-CNN framework |
| CN112327331A (en) * | 2020-11-02 | 2021-02-05 | 中山大学 | GNSS deception jamming detection method, device, equipment and storage medium |
| CN114092844A (en) * | 2021-11-19 | 2022-02-25 | 中国直升机设计研究所 | Multi-band image target detection method based on generation countermeasure network |
| CN115494526A (en) * | 2022-08-29 | 2022-12-20 | 中山大学 | GNSS deception jamming detection method and device, electronic equipment and storage medium |
| CN115598669A (en) * | 2022-09-26 | 2023-01-13 | 中山大学（Cn） | Navigation multi-feature GNSS deception jamming detection method, system, equipment and medium |
Non-Patent Citations (2)
| Title |
|---|
| JUNZHI LI 等: "GNSS Spoofing Jamming Detection Based on Generative Adversarial Network", IEEE SENSORS JOURNAL, vol. 21, no. 20, 15 October 2021 (2021-10-15), pages 22823 - 22832, XP011884414, DOI: 10.1109/JSEN.2021.3105404 * |
| 钟跃崎: "人工智能技术原理与应用", 30 September 2020, 东华大学出版社, pages: 238 - 239 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Abbas et al. | WiDeep: WiFi-based accurate and robust indoor localization system using deep learning | |
| Ding et al. | Target reconstruction based on 3-D scattering center model for robust SAR ATR | |
| CN110166991B (en) | Method, device, apparatus and storage medium for locating electronic device | |
| CN111680675B (en) | Face living body detection method, system, device, computer equipment and storage medium | |
| Krish et al. | Pre‐registration of latent fingerprints based on orientation field | |
| CN111722251A (en) | Weak signal capturing method and device, computer equipment and storage medium | |
| CN111953665B (en) | Server attack access identification method and system, computer equipment and storage medium | |
| CN105792131A (en) | Positioning method and system | |
| CN115598669A (en) | Navigation multi-feature GNSS deception jamming detection method, system, equipment and medium | |
| CN111047088A (en) | Prediction image acquisition method and device, computer equipment and storage medium | |
| CN116385745A (en) | Image recognition method, device, electronic equipment and storage medium | |
| CN113780363B (en) | Method, system, computer and medium for defending countermeasures | |
| CN111260024A (en) | Fault detection method and system based on combination of long-term and short-term memory and typical correlation | |
| CN111582378B (en) | Training generation method, position detection method and device of positioning recognition model | |
| CN113609097A (en) | Fingerprint library generation method and device, computer equipment and storage medium | |
| Zou et al. | A transfer kernel learning based strategy for adaptive localization in dynamic indoor environments: poster | |
| CN116736340A (en) | Deception signal detection method, deception signal detection device, computer equipment and storage medium | |
| KR101479384B1 (en) | Methods and apparatuses for facilitating object recognition | |
| CN111695377B (en) | Text detection method and device and computer equipment | |
| CN110471088B (en) | Positioning information processing method and device, computer equipment and storage medium | |
| CN111723686A (en) | Signal number estimation method and device, electronic equipment and storage medium | |
| CN112508789A (en) | Residual error-based patch image enhancement identification method and device | |
| CN112348060A (en) | Classification vector generation method and device, computer equipment and storage medium | |
| CN117421644B (en) | Wireless equipment intrusion detection method and device | |
| CN114358094B (en) | Signal denoising method and system based on radar communication system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |