CN111741471B - Intrusion detection method and device based on CSI and computer storage medium - Google Patents
Intrusion detection method and device based on CSI and computer storage medium Download PDFInfo
- Publication number
- CN111741471B CN111741471B CN202010533116.7A CN202010533116A CN111741471B CN 111741471 B CN111741471 B CN 111741471B CN 202010533116 A CN202010533116 A CN 202010533116A CN 111741471 B CN111741471 B CN 111741471B
- Authority
- CN
- China
- Prior art keywords
- amplitude
- phase difference
- data
- target
- mean square
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 51
- 239000013598 vector Substances 0.000 claims abstract description 135
- 238000000034 method Methods 0.000 claims abstract description 20
- 238000012549 training Methods 0.000 claims description 56
- 238000007781 pre-processing Methods 0.000 claims description 25
- 238000012545 processing Methods 0.000 claims description 24
- 238000001914 filtration Methods 0.000 claims description 17
- 238000000513 principal component analysis Methods 0.000 claims description 17
- 238000000605 extraction Methods 0.000 claims description 16
- 230000009545 invasion Effects 0.000 claims description 5
- 108091006146 Channels Proteins 0.000 description 81
- 230000006870 function Effects 0.000 description 24
- 238000005516 engineering process Methods 0.000 description 8
- 230000008569 process Effects 0.000 description 7
- 238000011160 research Methods 0.000 description 7
- 238000004891 communication Methods 0.000 description 6
- 230000002159 abnormal effect Effects 0.000 description 5
- 230000007704 transition Effects 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 238000012706 support-vector machine Methods 0.000 description 3
- 238000011161 development Methods 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- 230000006399 behavior Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000001035 drying Methods 0.000 description 1
- 230000003631 expected effect Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000005259 measurement Methods 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
- 238000012847 principal component analysis method Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W12/00—Security arrangements; Authentication; Protecting privacy or anonymity
- H04W12/12—Detection or prevention of fraud
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Computer Networks & Wireless Communication (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Signal Processing (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computer Security & Cryptography (AREA)
- Radar Systems Or Details Thereof (AREA)
Abstract
The invention discloses an intrusion detection method and device based on CSI and a computer storage medium, wherein the method comprises the following steps: acquiring channel state information data under a scene to be detected; extracting characteristics of the channel state information data to obtain amplitude characteristic vectors and phase difference characteristic vectors; inputting the amplitude characteristic vector and the phase difference characteristic vector into a target classifier, and outputting a corresponding class label by the target classifier; and traversing the category labels by using a time window to detect whether the scene to be detected is a scene with personnel intrusion. Therefore, by using the amplitude characteristic vector and the phase difference characteristic vector of the channel state information data as characteristic information, the accuracy of the personnel intrusion detection result based on the channel state information is improved.
Description
Technical Field
The present invention relates to the field of signal processing technologies, and in particular, to an intrusion detection method and apparatus based on CSI, and a computer storage medium.
Background
With the development of technology and the popularization of WIFI router equipment, the application of the WIFI wireless local area network technology in the field of personnel intrusion detection becomes the research direction of more and more students in China and abroad. Because equipment such as the router is cheap and easy to obtain, the technology has very high popularity, and personnel intrusion detection can be carried out only by means of collected WIFI signals without carrying any extra equipment by detection personnel for detection, so that the convenience and practicality of the technology are greatly improved.
At present, researchers commonly use channel state information (Chanel Status Information, CSI) in WIFI signals to detect personnel invasion, the channel state information has amplitude information and phase information, the amplitude information has better stability than the phase information, the phase information has finer granularity than the amplitude information in describing personnel disturbance, the existing personnel invasion detection research based on the channel state information processes the acquired original data through filtering, phase compensation, principal component analysis method dimension reduction and the like, and then classification training is carried out through a support vector machine algorithm, a Bayesian algorithm and the like. However, the intrusion detection system built in the actual test can accurately identify illegal intrusion behaviors, so that the expected effect is achieved, the robustness of the detection system is reduced in a complex environment, and the accuracy of the detection result is reduced in the complex environment.
Disclosure of Invention
The invention provides an intrusion detection method and device based on CSI and a computer storage medium, and aims to solve the technical problem that the accuracy of intrusion detection results based on CSI is low in the current complex scene.
In order to achieve the above object, the present invention provides a CSI-based intrusion detection method, including:
Acquiring channel state information data under a scene to be detected;
extracting characteristics of the channel state information data to obtain amplitude characteristic vectors and phase difference characteristic vectors;
inputting the amplitude characteristic vector and the phase difference characteristic vector into a target classifier, and outputting a corresponding class label by the target classifier;
and traversing the category labels by using a time window to detect whether the scene to be detected is a scene with personnel intrusion.
Preferably, amplitude data and phase data corresponding to the channel state information data are obtained;
preprocessing the amplitude data to obtain target amplitude data, and extracting features of the target amplitude data to obtain an amplitude feature vector;
preprocessing the phase data to obtain target phase difference data, and extracting features of the target phase difference data to obtain a phase difference feature vector.
Preferably, hampel filtering processing is carried out on the amplitude data so as to obtain target amplitude data;
dividing the target amplitude data into a preset number of short-sequence amplitude data by using a time window;
obtaining first-order differential amplitude data corresponding to the short-sequence amplitude data in each time window;
Performing principal component analysis on the first-order differential amplitude data to obtain a first amplitude principal component and a second amplitude principal component corresponding to the short-sequence amplitude data in each time window;
and acquiring an amplitude characteristic vector according to the first amplitude main component and the second amplitude main component.
Preferably, a first amplitude mean square error corresponding to the first amplitude principal component and a second amplitude mean square error corresponding to the second amplitude principal component are calculated respectively, so as to obtain an amplitude mean square error average value according to the first amplitude mean square error and the second amplitude mean square error;
and acquiring an amplitude characteristic vector based on the amplitude mean square error.
Preferably, the phase data are subjected to phase compensation and phase linearization in sequence to obtain linear phase data;
determining phase difference data corresponding to the linear phase data, and performing Hampel filtering processing on the phase difference data to obtain target phase difference data;
dividing the target phase difference data into a preset number of short-sequence phase difference data by using a time window;
obtaining first-order differential phase difference data corresponding to the short-sequence phase difference data in each time window;
performing principal component analysis on the first-order differential phase difference data to obtain a first phase difference principal component and a second phase difference principal component corresponding to the short-sequence phase difference data in each time window;
And obtaining a phase difference characteristic vector according to the first phase difference main component and the second phase difference main component.
Preferably, a first phase difference mean square error and a second phase difference mean square error corresponding to the first phase difference main component and the second phase difference main component are calculated respectively, so as to obtain a phase difference mean square error average value according to the first phase difference mean square error and the second phase difference mean square error;
and acquiring a phase difference characteristic vector based on the phase difference mean square error.
Preferably, traversing the class labels by using a time window, and judging whether a preset number of first class labels exist in the time window;
if the first type labels with the preset number exist in the time window, judging that the scene to be detected is a personnel intrusion scene.
Preferably, obtaining channel state information training data in a known state scene, and obtaining a prediction type label corresponding to the channel state information training data by using an initial classifier;
calculating a loss function corresponding to the initial classifier based on the prediction classification label and a real class label corresponding to the channel state information training data;
updating parameters of the initial classifier in a gradient descent manner based on the loss function;
And if the loss function reaches the convergence condition, stopping updating, and storing target parameters corresponding to the loss function reaching the convergence condition to obtain the target classifier.
In addition, in order to achieve the above object, the present invention also provides a CSI-based intrusion detection apparatus, including:
the acquisition module is used for acquiring channel state information data in a scene to be detected;
the extraction module is used for carrying out feature extraction on the channel state information data so as to obtain an amplitude feature vector and a phase difference feature vector;
the input module is used for inputting the amplitude characteristic vector and the phase difference characteristic vector into a target classifier, and outputting a corresponding class label by the target classifier;
and the detection module is used for traversing the category labels by utilizing a time window so as to detect whether the scene to be detected is a scene with personnel intrusion.
In addition, to achieve the above object, the present invention also provides a computer storage medium having stored thereon a CSI-based intrusion detection program which, when executed by a processor, implements the steps of the CSI-based intrusion detection method as described above.
Compared with the prior art, the invention discloses an intrusion detection method, an intrusion detection device and a computer storage medium based on CSI, which are used for acquiring channel state information data under a scene to be detected, carrying out feature extraction on the channel state information data to acquire an amplitude feature vector and a phase difference feature vector, inputting the amplitude feature vector and the phase difference feature vector into a target classifier, outputting corresponding class labels by the target classifier, and traversing the class labels by utilizing a time window to detect whether the scene to be detected is a scene with personnel intrusion. Therefore, by using the amplitude characteristic vector and the phase difference characteristic vector of the channel state information data as characteristic information, the accuracy of the personnel intrusion detection result based on the channel state information is improved.
Drawings
FIG. 1 is a schematic diagram of a terminal architecture of a hardware operating environment according to embodiments of the present invention;
fig. 2 is a flowchart of a first embodiment of the CSI-based intrusion detection method according to the present invention;
FIG. 3 is a flow chart of a second embodiment of the CSI-based intrusion detection method of the present invention;
fig. 4 is a functional block diagram of a first embodiment of the CSI-based intrusion detection device according to the present invention.
The achievement of the objects, functional features and advantages of the present invention will be further described with reference to the accompanying drawings, in conjunction with the embodiments.
Detailed Description
It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
Referring to fig. 1, fig. 1 is a schematic diagram of a terminal structure of a hardware operating environment according to embodiments of the present invention. In an embodiment of the present invention, the terminal device may include a processor 1001 (e.g., a central processing unit Central Processing Unit, a CPU), a communication bus 1002, an input port 1003, an output port 1004, and a memory 1005. Wherein the communication bus 1002 is used to enable connected communications between these components; the input port 1003 is used for data input; the output port 1004 is used for data output, and the memory 1005 may be a high-speed RAM memory or a stable memory (non-volatile memory), such as a disk memory, and the memory 1005 may be an optional storage device independent of the processor 1001. Those skilled in the art will appreciate that the hardware configuration shown in fig. 1 is not limiting of the invention and may include more or fewer components than shown, or may combine certain components, or a different arrangement of components.
With continued reference to fig. 1, the memory 1005 of fig. 1, which is a readable storage medium, may include an operating system, a network communication module, an application module, and a CSI-based intrusion detection program. In fig. 1, the network communication module is mainly used for connecting with a server and performing data communication with the server; and the processor 1001 may call the CSI-based intrusion detection program stored in the memory 1005 and perform the following operations:
acquiring channel state information data under a scene to be detected;
extracting characteristics of the channel state information data to obtain amplitude characteristic vectors and phase difference characteristic vectors;
inputting the amplitude characteristic vector and the phase difference characteristic vector into a target classifier, and outputting a corresponding class label by the target classifier;
and traversing the category labels by using a time window to detect whether the scene to be detected is a scene with personnel intrusion.
Further, the processor 1001 may be further configured to invoke the CSI-based intrusion detection program stored in the memory 1005, and perform the following steps:
acquiring amplitude data and phase data corresponding to the channel state information data;
preprocessing the amplitude data to obtain target amplitude data, and extracting features of the target amplitude data to obtain an amplitude feature vector;
Preprocessing the phase data to obtain target phase difference data, and extracting features of the target phase difference data to obtain a phase difference feature vector.
Further, the processor 1001 may be further configured to invoke the CSI-based intrusion detection program stored in the memory 1005, and perform the following steps:
hampel filtering is carried out on the amplitude data to obtain target amplitude data;
dividing the target amplitude data into a preset number of short-sequence amplitude data by using a time window;
obtaining first-order differential amplitude data corresponding to the short-sequence amplitude data in each time window;
performing principal component analysis on the first-order differential amplitude data to obtain a first amplitude principal component and a second amplitude principal component corresponding to the short-sequence amplitude data in each time window;
and acquiring an amplitude characteristic vector according to the first amplitude main component and the second amplitude main component.
Further, the processor 1001 may be further configured to invoke the CSI-based intrusion detection program stored in the memory 1005, and perform the following steps:
respectively calculating a first amplitude mean square error corresponding to the first amplitude principal component and a second amplitude mean square error corresponding to the second amplitude principal component so as to obtain an amplitude mean square error average value according to the first amplitude mean square error and the second amplitude mean square error;
And acquiring an amplitude characteristic vector based on the amplitude mean square error.
Further, the processor 1001 may be further configured to invoke the CSI-based intrusion detection program stored in the memory 1005, and perform the following steps:
sequentially performing phase compensation and phase linearization on the phase data to obtain linear phase data;
determining phase difference data corresponding to the linear phase data, and performing Hampel filtering processing on the phase difference data to obtain target phase difference data;
dividing the target phase difference data into a preset number of short-sequence phase difference data by using a time window;
obtaining first-order differential phase difference data corresponding to the short-sequence phase difference data in each time window;
performing principal component analysis on the first-order differential phase difference data to obtain a first phase difference principal component and a second phase difference principal component corresponding to the short-sequence phase difference data in each time window;
and obtaining a phase difference characteristic vector according to the first phase difference main component and the second phase difference main component.
Further, the processor 1001 may be further configured to invoke the CSI-based intrusion detection program stored in the memory 1005, and perform the following steps:
Respectively calculating a first phase difference mean square error corresponding to the first phase difference main component and a second phase difference mean square error corresponding to the second phase difference main component so as to obtain a phase difference mean square error average value according to the first phase difference mean square error and the second phase difference mean square error;
and acquiring a phase difference characteristic vector based on the phase difference mean square error.
Further, the processor 1001 may be further configured to invoke the CSI-based intrusion detection program stored in the memory 1005, and perform the following steps:
traversing the category labels by using a time window, and judging whether a preset number of first category labels exist in the time window;
if the first type labels with the preset number exist in the time window, judging that the scene to be detected is a personnel intrusion scene.
Further, the processor 1001 may be further configured to invoke the CSI-based intrusion detection program stored in the memory 1005, and perform the following steps:
acquiring channel state information training data under a known state scene, and acquiring a prediction category label corresponding to the channel state information training data by using an initial classifier;
calculating a loss function corresponding to the initial classifier based on the prediction classification label and a real class label corresponding to the channel state information training data;
Updating parameters of the initial classifier in a gradient descent manner based on the loss function;
and if the loss function reaches the convergence condition, stopping updating, and storing target parameters corresponding to the loss function reaching the convergence condition to obtain the target classifier.
Based on the above structure, various embodiments of the intrusion detection method based on CSI of the present invention are presented.
Referring to fig. 2, fig. 2 is a flowchart illustrating a first embodiment of the CSI-based intrusion detection method according to the present invention.
In this embodiment, the intrusion detection method based on CSI includes:
step S10: acquiring channel state information data under a scene to be detected;
in this embodiment, with the development of technology and popularization of WIFI router devices, WIFI wireless lan technology is applied to the field of personnel intrusion detection to become a research direction of more and more students in China and abroad, because the router and other devices are cheap and easy to obtain, the WIFI router device has very high popularity, personnel intrusion detection can be performed only by means of collected WIFI signals without carrying any additional devices by detection personnel for detection, convenience and practicality of the technology are greatly improved, most researchers currently adopt wireless signal receiving intensity indication parameters (Received Signal Strength Indication, RSSI) as research parameters of personnel intrusion detection, however, only one RSSI information in one data packet causes a larger research difficulty, so in this embodiment, wireless channel state information (Chanel Status Information, CSI) is adopted to replace the traditional wireless signal receiving intensity indication parameters, the problem that signal power is dependent in the traditional method can be solved, in particular, the personnel intrusion can affect a wireless link, and is reflected on amplitude and phase fluctuation in CSI information, so that whether personnel intrusion exists can be detected through amplitude and phase.
In a specific application, channel state information data received by multiple antennas at a receiving end is obtained, for example, channel state information data corresponding to three antennas at a receiving end of a wireless signal under a scene to be detected is obtained, wherein a transmitting end for transmitting the wireless signal and a receiving end for receiving the wireless signal are both common commercial devices, for example, a router can be used as a transmitter, a notebook can be used as a receiver, but the router is not limited to the two devices, optionally, 30 subcarrier clusters of data under the scene to be detected are obtained by using an Intel5300 network card, and then 30 channel state information data corresponding to the 30 subcarrier clusters of data are obtained.
Step S20: extracting characteristics of the channel state information data to obtain amplitude characteristic vectors and phase difference characteristic vectors;
the channel state information includes two kinds of information, namely amplitude and phase, wherein the amplitude information has better stability than the phase information, but the phase information has finer granularity than the amplitude information in the disturbance of a constructor, so in the embodiment, the two kinds of information, namely the amplitude information and the phase corresponding to the channel state information, are extracted as characteristic information for research.
It should be noted that, the original channel state information data collected from the receiving end antenna cannot be directly used for personnel intrusion recognition, because there are many interferences, such as environmental noise, electrical noise and other nearby WIFI signal interferences, these interference factors may cause severe abnormal fluctuation of the original channel state information data collected by us, so as to affect the discrimination of our intrusion detection, therefore we need to perform preprocessing such as filtering on the collected data information to remove noise points and abnormal data points of the original channel state information data, and perform feature extraction on the preprocessed channel state information data after preprocessing.
It should be noted that, since the phase difference often has a larger variance, the embodiment uses the phase difference as the feature information, specifically, extracts the amplitude data and the phase data corresponding to the channel state information data, then calculates the phase difference data corresponding to the three antennas at the receiving end, and finally performs feature extraction on the amplitude data and the phase difference data to obtain the amplitude feature vector and the phase difference feature vector.
Step S30: inputting the amplitude characteristic vector and the phase difference characteristic vector into a target classifier, and outputting a corresponding class label by the target classifier;
step S40: and traversing the category labels by using a time window to detect whether the scene to be detected is a scene with personnel intrusion.
In this embodiment, the classifier may select a support vector machine (Support Vector Machine, SVM) to output class labels corresponding to the amplitude feature vector and the phase difference feature vector, and further, before the SVM classifier is used to perform feature vector on the amplitude feature vector and the phase difference feature vector input each time, training data is further required to train the SVM classifier to obtain a trained target classifier, so as to output the class labels corresponding to the amplitude feature vector and the phase difference feature vector input each time based on the target classifier.
Specifically, before the step of inputting the amplitude feature vector and the phase difference feature vector into the target classifier, the method further includes:
step a: acquiring channel state information training data under a known state scene, and acquiring a prediction category label corresponding to the channel state information training data by using an initial classifier;
step b: calculating a loss function corresponding to the initial classifier based on the prediction classification label and a real class label corresponding to the channel state information training data;
step c: updating parameters of the initial classifier in a gradient descent manner based on the loss function;
step d: and if the loss function reaches the convergence condition, stopping updating, and storing target parameters corresponding to the loss function reaching the convergence condition to obtain the target classifier.
In the step, a large amount of training data is randomly prepared, real class labels corresponding to the training data are calibrated, the real class labels are determined according to the real scene people number state of the training data, for example, first channel state information training data in a person scene are randomly acquired, the class labels of the first channel state information training data are marked as first class labels, for example, 1, wherein the first channel state information training data in the person scene can be channel state information training data corresponding to the same scene but different people, or channel state information training data corresponding to different scenes and different people, the training data are not limited, further, in order to guarantee the accuracy of classification of a target classifier, the training data also comprise second channel state information training data in an unmanned scene, the class labels of the first channel state information training data are marked as second class labels, for example, 0, and then preprocessing and feature extraction processing are carried out on the first channel state information training data and the second channel state information training data respectively.
For example, taking the first channel state information training data as an example, extracting first amplitude training data and first phase training data corresponding to the first channel state information training data, performing phase compensation and linearization on the first phase training data to obtain first linear phase training data corresponding to the first phase training data, then obtaining first phase difference training data corresponding to the first linear phase training data, and finally performing Hampel filtering on the first amplitude training data and the first phase difference training data to perform denoising processing on the first amplitude training data and the first phase difference training data. The second channel state information training data processing step is similar to the first channel state information training data, and after the first amplitude training data, the first phase difference training data, the second amplitude training data and the second phase difference training data are obtained based on the steps, feature extraction is performed on the first amplitude training data, the first phase difference training data, the second amplitude training data and the second phase difference training data respectively, so as to obtain corresponding first amplitude feature vectors, first phase difference feature vectors, second amplitude feature vectors and second phase difference feature vectors.
Then, the first amplitude feature vector, the first phase difference feature vector, the second amplitude feature vector and the second phase difference feature vector are used as training data to train an initial classifier, the first amplitude feature vector and the first phase difference feature vector are taken as examples, the first amplitude feature vector and the first phase difference feature vector are input into the initial classifier, such as the initial SVM classifier, so that the initial SVM classifier outputs a corresponding prediction class label, then a loss function corresponding to the initial SVM classifier is calculated according to the prediction class label and a corresponding real class label (namely the first class label), and optionally, a mini-batch is adopted to calculate a cross entropy loss function of the initial SVM classifier.
After the cross entropy loss function of the initial SVM classifier is obtained, gradients corresponding to all parameters in the initial SVM classifier are calculated according to the cross entropy loss function, and all parameters are correspondingly updated according to the gradients of all the parameters, namely all the parameters of the initial SVM classifier are adjusted. Here, the process of updating the model parameters according to the cross entropy loss function is similar to the existing model parameter updating process, and detailed description is omitted herein until the cross entropy loss function reaches the convergence condition, updating of each parameter in the initial SVM classifier is stopped, and each corresponding parameter value when the cross entropy loss function reaches the convergence condition is confirmed and stored, so as to obtain the target classifier.
It should be further noted that, in this embodiment, the channel state information data is divided into a preset number of short-sequence channel state information data by using a time window, that is, the amplitude feature vector and the phase difference feature vector corresponding to the preset number are obtained to improve the processing efficiency of the data, so that the preset number of amplitude feature vectors and the phase difference feature vector are input into the target classifier one by one to obtain the class labels corresponding to the preset number, and then the preset number of class labels are traversed by using the time window for the second time to detect whether the scene to be detected is a human intrusion scene.
Specifically, the step of traversing the category labels by using a time window to detect whether the scene to be detected is a person-intrusion scene includes:
step d: traversing the category labels by using a time window, and judging whether a preset number of first category labels exist in the time window;
step e: if the first type labels with the preset number exist in the time window, judging that the scene to be detected is a personnel intrusion scene.
In this step, since the first channel state information data in the manned scene is labeled with the first type label during training in this embodiment to train the SVM classifier to obtain the target classifier, the first type label output by the target classifier may be the feature corresponding to the channel state information data in the manned scene, specifically, the preset number of amplitude feature vectors and phase difference feature vectors are traversed by using the time window to one-to-one correspond to the preset number of type labels, and whether the preset number of first type labels exist in the time window is determined, for example, if 5 times of 1 (i.e., the first type labels) appear in the window in an accumulated manner, the scene to be detected is determined to be invaded by a person, i.e., the alert device of the scene to be detected may be triggered at this time to prompt the manager of the invasion of the person.
According to the scheme, the channel state information data under the scene to be detected is obtained, the channel state information data is subjected to feature extraction to obtain the amplitude feature vector and the phase difference feature vector, the amplitude feature vector and the phase difference feature vector are input into the target classifier, the target classifier outputs corresponding class labels, the class labels are traversed through a time window to detect whether the scene to be detected is a person intrusion scene, and therefore the accuracy of the person intrusion detection result based on the channel state information is improved by using the amplitude feature vector and the phase difference feature vector of the channel state information data as feature information.
Based on the second embodiment shown in fig. 2 described above, a third embodiment of the present invention is proposed. As shown in fig. 3, fig. 3 is a flowchart illustrating a second embodiment of the CSI-based intrusion detection method according to the present invention.
The step of extracting the characteristics of the channel state information data to obtain an amplitude characteristic vector and a phase difference characteristic vector comprises the following steps:
step S201: acquiring amplitude data and phase data corresponding to the channel state information data;
it is understood that the channel state information includes both amplitude and phase information, and thus, after the channel state information data is acquired, amplitude data and phase data corresponding to the channel state information data are extracted.
Step S202: preprocessing the amplitude data to obtain target amplitude data, and extracting features of the target amplitude data to obtain an amplitude feature vector;
it should be noted that, the original channel state information data collected from the receiving end antenna cannot be directly used for personnel intrusion recognition, because there are many interferences, such as environmental noise, electrical noise and other nearby WIFI signal interferences, and these interference factors may cause severe abnormal fluctuation of the original channel state information data collected by us, so as to affect the discrimination of our intrusion detection, so we need to perform preprocessing such as filtering on the collected data information, that is, after removing the noise points and abnormal data points of the original channel state information data, perform feature extraction on the data to obtain feature vectors.
Further, since the channel state information includes two kinds of information of amplitude and phase, and there is a certain difference between the two kinds of information of amplitude and phase, after the amplitude information and the phase information corresponding to the channel state information are extracted, the amplitude information and the phase information are processed respectively, so that the accuracy of data is improved, and the detection result is further improved.
Specifically, step S202 includes:
step S202a: hampel filtering is carried out on the amplitude data to obtain target amplitude data;
step S202b: dividing the target amplitude data into a preset number of short-sequence amplitude data by using a time window;
step S202c: obtaining first-order differential amplitude data corresponding to the short-sequence amplitude data in each time window;
step S202d: performing principal component analysis on the first-order differential amplitude data to obtain a first amplitude principal component and a second amplitude principal component corresponding to the short-sequence amplitude data in each time window;
step S202e: and acquiring an amplitude characteristic vector according to the first amplitude main component and the second amplitude main component.
In this step, after the amplitude data is obtained, hampel filtering processing is performed on the amplitude data to delete the abnormal amplitude data and obtain the target amplitude data, further, research finds that curve fluctuation after first-order difference of the corresponding channel state information data is more severe than curve fluctuation before first-order difference when someone invades, that is, the first-order difference can amplify influence on the channel state information data when someone invades so as to be detected and judge invasion, so that the first-order difference processing is performed on the amplitude data, further, in order to improve data processing efficiency, in this embodiment, the target amplitude data is firstly divided into short-sequence amplitude data with a preset number by using a time window, then, first-order difference amplitude data corresponding to the short-sequence amplitude data in each time window is respectively obtained at the same time, and then principal component analysis dimension reduction is performed on the short-sequence amplitude data in each time window to obtain a first amplitude principal component and a second amplitude principal component, and finally, an amplitude feature vector is obtained according to the first amplitude principal component and the second amplitude principal component.
Specifically, the step of obtaining the amplitude feature vector according to the first amplitude principal component and the second amplitude principal component includes:
respectively calculating a first amplitude mean square error corresponding to the first amplitude principal component and a second amplitude mean square error corresponding to the second amplitude principal component so as to obtain an amplitude mean square error average value according to the first amplitude mean square error and the second amplitude mean square error;
and acquiring an amplitude characteristic vector based on the amplitude mean square error.
In the step, the first amplitude principal component corresponding to the first principal component first order difference amplitude data and the second amplitude principal component corresponding to the second principal component first order difference amplitude data are determined, then the mean square error is respectively calculated, and finally the mean value between the first amplitude principal component mean square error and the first amplitude principal component mean square error is calculated to obtain an amplitude characteristic vector.
Step S203: preprocessing the phase data to obtain target phase difference data, and extracting features of the target phase difference data to obtain a phase difference feature vector.
In this step, in order to avoid the influence of the drying factor on the detection result, the phase data is preprocessed to obtain the target phase difference data, and then the feature extraction is performed on the target phase difference data to obtain the phase difference feature vector.
Specifically, step S203 includes:
step S203a: sequentially performing phase compensation and phase linearization on the phase data to obtain linear phase data;
step S203b: determining phase difference data corresponding to the linear phase data, and performing Hampel filtering processing on the phase difference data to obtain target phase difference data;
step S203c: dividing the target phase difference data into a preset number of short-sequence phase difference data by using a time window;
step S203d: obtaining first-order differential phase difference data corresponding to the short-sequence phase difference data in each time window;
step S203e: performing principal component analysis on the first-order differential phase difference data to obtain a first phase difference principal component and a second phase difference principal component corresponding to the short-sequence phase difference data in each time window;
step S203f: and obtaining a phase difference characteristic vector according to the first phase difference main component and the second phase difference main component.
In this step, since the random distribution of the measurement phase pairs is obtained from the Intel5300 network card, the CSI original phase data curve in the data packet is distributed in disorder, so that the disturbance to the wireless network environment generated when the personnel invade is detected cannot be detected through the disorder curve distribution, that is, the personnel cannot be perceived and invaded to be detected by using the phase information, therefore, in this embodiment, after the phase data is extracted, the phase data is subjected to the phase compensation processing to obtain the complete phase data, wherein the transition directions of the original CSI phase are respectively from-pi transition to +pi transition to-pi transition, and the CSI phase is compensated differently according to different phase transition directions, wherein the compensated phase is 2 pi or-2 pi.
Further, since the phase difference tends to have a larger variance, the present embodiment uses the phase difference as the characteristic information, so in the present embodiment, in order to conveniently and accurately acquire the phase difference characteristic information, after performing phase compensation, phase linearization processing is performed on the phase data to acquire linear phase data.
Further, in order to obtain the characteristic with larger difference and improve the data processing efficiency, in this embodiment, after obtaining the target phase difference data, the target phase difference data is first divided into a preset number of short-sequence phase difference data by using a time window, then the preset number of short-sequence phase difference data is subjected to first-order differential processing, and finally the phase difference feature vector is obtained based on the phase difference data after the first-order differential processing, which is not described herein.
Further, the step of obtaining the phase difference feature vector according to the first phase difference principal component and the second phase difference principal component includes:
respectively calculating a first phase difference mean square error corresponding to the first phase difference main component and a second phase difference mean square error corresponding to the second phase difference main component so as to obtain a phase difference mean square error average value according to the first phase difference mean square error and the second phase difference mean square error;
And acquiring a phase difference characteristic vector based on the phase difference mean square error.
In the step, the first phase difference main component corresponding to the first main component first-order difference phase difference data and the second phase difference main component corresponding to the second main component first-order difference phase difference data are determined, then the mean square error of the first phase difference main component and the mean square error of the first phase difference main component are respectively calculated, and finally the phase difference characteristic vector is obtained.
According to the embodiment, through the scheme, amplitude data and phase data corresponding to the channel state information data are obtained; preprocessing the amplitude data to obtain target amplitude data, and extracting features of the target amplitude data to obtain an amplitude feature vector; the phase data is preprocessed to obtain target phase difference data, and the target phase difference data is subjected to feature extraction to obtain a phase difference feature vector, so that the accuracy of a personnel intrusion detection result based on channel state information is improved by using the amplitude feature vector and the phase difference feature vector of the channel state information data as feature information.
In addition, the embodiment also provides an intrusion detection device based on the CSI. Referring to fig. 4, fig. 4 is a schematic functional block diagram of a first embodiment of an intrusion detection device based on CSI according to the present invention.
In this embodiment, the intrusion detection device based on CSI is a virtual device, and is stored in the memory 1005 of the terminal device shown in fig. 1, so as to implement all functions of the intrusion detection program based on CSI: the method comprises the steps of acquiring channel state information data under a scene to be detected; the method comprises the steps of extracting characteristics of channel state information data to obtain amplitude characteristic vectors and phase difference characteristic vectors; the amplitude characteristic vector and the phase difference characteristic vector are input into a target classifier, and the target classifier outputs corresponding class labels; and traversing the category labels by using a time window to detect whether the scene to be detected is a scene with personnel intrusion.
Specifically, the CSI-based intrusion detection apparatus includes:
an acquisition module 10, configured to acquire channel state information data in a scene to be detected;
the extracting module 20 is configured to perform feature extraction on the channel state information data to obtain an amplitude feature vector and a phase difference feature vector;
the input module 30 is configured to input the amplitude feature vector and the phase difference feature vector to a target classifier, and output a corresponding class label by the target classifier;
the detection module 40 is configured to traverse the category label by using a time window to detect whether the scene to be detected is a scene having a person intruding.
Further, the extraction module includes:
the first acquisition unit is used for acquiring amplitude data and phase data corresponding to the channel state information data;
the second acquisition unit is used for preprocessing the amplitude data to acquire target amplitude data, and extracting features of the target amplitude data to acquire an amplitude feature vector;
and the third acquisition unit is used for preprocessing the phase data to acquire target phase difference data and extracting characteristics of the target phase difference data to acquire a phase difference characteristic vector.
Further, the second obtaining unit further includes:
the first preprocessing subunit is used for carrying out Hampel filtering processing on the amplitude data so as to acquire target amplitude data;
a first dividing subunit, configured to divide the target amplitude data into a preset number of short-sequence amplitude data by using a time window;
the first solving subunit is used for solving first-order differential amplitude data corresponding to the short-sequence amplitude data in each time window;
the first principal component analysis subunit is used for performing principal component analysis on the first-order differential amplitude data to obtain a first amplitude principal component and a second amplitude principal component corresponding to the short-sequence amplitude data in each time window;
The first acquisition subunit is configured to acquire an amplitude feature vector according to the first amplitude principal component and the second amplitude principal component.
Further, the first acquisition subunit is further configured to:
respectively calculating a first amplitude mean square error corresponding to the first amplitude principal component and a second amplitude mean square error corresponding to the second amplitude principal component so as to obtain an amplitude mean square error average value according to the first amplitude mean square error and the second amplitude mean square error;
and acquiring an amplitude characteristic vector based on the amplitude mean square error.
Further, the third obtaining unit further includes:
the second preprocessing subunit is used for sequentially carrying out phase compensation and phase linearization on the phase data so as to obtain linear phase data;
the third preprocessing subunit is used for determining phase difference data corresponding to the linear phase data and carrying out Hampel filtering processing on the phase difference data to obtain target phase difference data;
a second dividing subunit for dividing the target phase difference data into a preset number of short-sequence phase difference data by using a time window;
the second calculating subunit is used for calculating first-order differential phase difference data corresponding to the short-sequence phase difference data in each time window;
The second principal component analysis subunit is used for performing principal component analysis on the first-order differential phase difference data to obtain a first phase difference principal component and a second phase difference principal component corresponding to the short-sequence phase difference data in each time window;
and the second acquisition subunit is used for acquiring the phase difference characteristic vector according to the first phase difference main component and the second phase difference main component.
Further, the second acquisition subunit is further configured to:
respectively calculating a first phase difference mean square error corresponding to the first phase difference main component and a second phase difference mean square error corresponding to the second phase difference main component so as to obtain a phase difference mean square error average value according to the first phase difference mean square error and the second phase difference mean square error;
and acquiring a phase difference characteristic vector based on the phase difference mean square error.
Further, the detection module further includes:
the judging unit is used for traversing the class labels by utilizing the time window and judging whether the first class labels with the preset number exist in the time window or not;
and the judging unit is used for judging that the scene to be detected is a personnel intrusion scene if the first type tags with the preset number exist in the time window.
Further, the input module further includes:
a fourth obtaining unit, configured to obtain channel state information training data in a known state scene, and obtain a prediction class label corresponding to the channel state information training data by using an initial classifier; the method comprises the steps of carrying out a first treatment on the surface of the
The calculating unit is used for calculating a loss function corresponding to the initial classifier based on the prediction classification label and the real class label corresponding to the channel state information training data;
an updating unit for updating parameters of the initial classifier in a gradient descent manner based on the loss function;
and a fifth obtaining unit, configured to stop updating if the loss function reaches a convergence condition, and store a target parameter corresponding to the loss function reaching the convergence condition, so as to obtain a target classifier.
In addition, the embodiment of the present invention further provides a computer storage medium, where a CSI-based intrusion detection program is stored, and when the CSI-based intrusion detection program is run by a processor, the steps of the CSI-based intrusion detection method described above are implemented, which is not described herein again.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or system. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or system that comprises the element.
The foregoing embodiment numbers of the present invention are merely for the purpose of description, and do not represent the advantages or disadvantages of the embodiments.
From the above description of the embodiments, it will be clear to those skilled in the art that the above-described embodiment method may be implemented by means of software plus a necessary general hardware platform, but of course may also be implemented by means of hardware, but in many cases the former is a preferred embodiment. Based on such understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art in the form of a software product stored in a storage medium (e.g. ROM/RAM, magnetic disk, optical disk) as described above, comprising several instructions for causing a terminal device to perform the method according to the embodiments of the present invention.
The foregoing description is only of the preferred embodiments of the present invention and is not intended to limit the scope of the invention, and all equivalent structures or modifications in the structures or processes described in the specification and drawings, or the direct or indirect application of the present invention to other related technical fields, are included in the scope of the present invention.
Claims (5)
1. The intrusion detection method based on the CSI is characterized by comprising the following steps of:
Acquiring channel state information data under a scene to be detected;
extracting characteristics of the channel state information data to obtain amplitude characteristic vectors and phase difference characteristic vectors;
inputting the amplitude characteristic vector and the phase difference characteristic vector into a target classifier, and outputting a corresponding class label by the target classifier;
traversing the category labels by using a time window to detect whether the scene to be detected is a scene with personnel invasion;
the step of extracting the characteristics of the channel state information data to obtain an amplitude characteristic vector and a phase difference characteristic vector comprises the following steps:
acquiring amplitude data and phase data corresponding to the channel state information data;
preprocessing the amplitude data to obtain target amplitude data, and extracting features of the target amplitude data to obtain an amplitude feature vector;
preprocessing the phase data to obtain target phase difference data, and extracting characteristics of the target phase difference data to obtain a phase difference characteristic vector;
the step of preprocessing the amplitude data to obtain target amplitude data and extracting features of the target amplitude data to obtain an amplitude feature vector comprises the following steps:
Hampel filtering is carried out on the amplitude data to obtain target amplitude data;
dividing the target amplitude data into a preset number of short-sequence amplitude data by using a time window;
obtaining first-order differential amplitude data corresponding to the short-sequence amplitude data in each time window;
performing principal component analysis on the first-order differential amplitude data to obtain a first amplitude principal component and a second amplitude principal component corresponding to the short-sequence amplitude data in each time window;
acquiring an amplitude characteristic vector according to the first amplitude main component and the second amplitude main component;
the step of obtaining the amplitude characteristic vector according to the first amplitude principal component and the second amplitude principal component comprises the following steps:
respectively calculating a first amplitude mean square error corresponding to the first amplitude principal component and a second amplitude mean square error corresponding to the second amplitude principal component so as to obtain an amplitude mean square error average value according to the first amplitude mean square error and the second amplitude mean square error;
acquiring an amplitude characteristic vector based on the amplitude mean square error;
the step of preprocessing the phase data to obtain target phase difference data and extracting features of the target phase difference data to obtain a phase difference feature vector comprises the following steps:
Sequentially performing phase compensation and phase linearization on the phase data to obtain linear phase data;
determining phase difference data corresponding to the linear phase data, and performing Hampel filtering processing on the phase difference data to obtain target phase difference data;
dividing the target phase difference data into a preset number of short-sequence phase difference data by using a time window;
obtaining first-order differential phase difference data corresponding to the short-sequence phase difference data in each time window;
performing principal component analysis on the first-order differential phase difference data to obtain a first phase difference principal component and a second phase difference principal component corresponding to the short-sequence phase difference data in each time window;
acquiring a phase difference characteristic vector according to the first phase difference main component and the second phase difference main component;
the step of obtaining the phase difference feature vector according to the first phase difference principal component and the second phase difference principal component includes:
respectively calculating a first phase difference mean square error corresponding to the first phase difference main component and a second phase difference mean square error corresponding to the second phase difference main component so as to obtain a phase difference mean square error average value according to the first phase difference mean square error and the second phase difference mean square error;
And acquiring a phase difference characteristic vector based on the phase difference mean square error.
2. The method of claim 1, wherein traversing the category labels with a time window to detect whether the scene to be detected is a person-intrusive scene comprises:
traversing the category labels by using a time window, and judging whether a preset number of first category labels exist in the time window;
if the first type labels with the preset number exist in the time window, judging that the scene to be detected is a personnel intrusion scene.
3. The method according to claim 1 or 2, further comprising, prior to the step of inputting the magnitude and phase difference eigenvectors to a target classifier:
acquiring channel state information training data under a known state scene, and acquiring a prediction category label corresponding to the channel state information training data by using an initial classifier;
calculating a loss function corresponding to the initial classifier based on the prediction class label and a real class label corresponding to the channel state information training data;
updating parameters of the initial classifier in a gradient descent manner based on the loss function;
And if the loss function reaches the convergence condition, stopping updating, and storing target parameters corresponding to the loss function reaching the convergence condition to obtain the target classifier.
4. A CSI-based intrusion detection apparatus, comprising:
the acquisition module is used for acquiring channel state information data in a scene to be detected;
the extraction module is used for carrying out feature extraction on the channel state information data so as to obtain an amplitude feature vector and a phase difference feature vector;
the input module is used for inputting the amplitude characteristic vector and the phase difference characteristic vector into a target classifier, and outputting a corresponding class label by the target classifier;
the detection module is used for traversing the category labels by utilizing a time window so as to detect whether the scene to be detected is a scene with personnel intruding;
the extraction module comprises:
the first acquisition unit is used for acquiring amplitude data and phase data corresponding to the channel state information data;
the second acquisition unit is used for preprocessing the amplitude data to acquire target amplitude data, and extracting features of the target amplitude data to acquire an amplitude feature vector;
The third acquisition unit is used for preprocessing the phase data to acquire target phase difference data, and extracting characteristics of the target phase difference data to acquire a phase difference characteristic vector;
the second acquisition unit includes:
the first preprocessing subunit is used for carrying out Hampel filtering processing on the amplitude data so as to acquire target amplitude data;
a first dividing subunit, configured to divide the target amplitude data into a preset number of short-sequence amplitude data by using a time window;
the first solving subunit is used for solving first-order differential amplitude data corresponding to the short-sequence amplitude data in each time window;
the first principal component analysis subunit is used for performing principal component analysis on the first-order differential amplitude data to obtain a first amplitude principal component and a second amplitude principal component corresponding to the short-sequence amplitude data in each time window;
the first acquisition subunit is used for acquiring an amplitude characteristic vector according to the first amplitude main component and the second amplitude main component;
the first obtaining subunit is further configured to calculate a first amplitude mean square error corresponding to the first amplitude principal component and a second amplitude mean square error corresponding to the second amplitude principal component, so as to obtain an amplitude mean square error average according to the first amplitude mean square error and the second amplitude mean square error; and acquiring an amplitude characteristic vector based on the amplitude mean square error.
The third obtaining unit further includes:
the second preprocessing subunit is used for sequentially carrying out phase compensation and phase linearization on the phase data so as to obtain linear phase data;
the third preprocessing subunit is used for determining phase difference data corresponding to the linear phase data and carrying out Hampel filtering processing on the phase difference data to obtain target phase difference data;
a second dividing subunit for dividing the target phase difference data into a preset number of short-sequence phase difference data by using a time window;
the second calculating subunit is used for calculating first-order differential phase difference data corresponding to the short-sequence phase difference data in each time window;
the second principal component analysis subunit is used for performing principal component analysis on the first-order differential phase difference data to obtain a first phase difference principal component and a second phase difference principal component corresponding to the short-sequence phase difference data in each time window;
a second obtaining subunit, configured to obtain a phase difference feature vector according to the first phase difference main component and the second phase difference main component;
the second obtaining subunit is further configured to calculate a first phase difference mean square error corresponding to the first phase difference main component and a second phase difference mean square error corresponding to the second phase difference main component, so as to obtain a phase difference mean square error average value according to the first phase difference mean square error and the second phase difference mean square error; and acquiring a phase difference characteristic vector based on the phase difference mean square error.
5. A computer storage medium, having stored thereon a CSI-based intrusion detection program, which when executed by a processor, implements the steps of the CSI-based intrusion detection method according to any one of claims 1 to 3.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010533116.7A CN111741471B (en) | 2020-06-11 | 2020-06-11 | Intrusion detection method and device based on CSI and computer storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010533116.7A CN111741471B (en) | 2020-06-11 | 2020-06-11 | Intrusion detection method and device based on CSI and computer storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111741471A CN111741471A (en) | 2020-10-02 |
| CN111741471B true CN111741471B (en) | 2024-03-22 |
Family
ID=72650148
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010533116.7A Active CN111741471B (en) | 2020-06-11 | 2020-06-11 | Intrusion detection method and device based on CSI and computer storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111741471B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112953663A (en) * | 2021-03-11 | 2021-06-11 | 南京信息工程大学 | Passive indoor intrusion detection method based on OFDM subcarrier empirical analysis |
| CN114359738B (en) * | 2022-03-16 | 2022-06-14 | 深圳大学 | Cross-scene robust indoor people number wireless detection method and system |
| CN115134815B (en) * | 2022-06-08 | 2024-02-09 | 深圳市三旺通信股份有限公司 | Wireless AP deployment optimization method, system, equipment and storage medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108199757A (en) * | 2018-01-15 | 2018-06-22 | 浙江大学 | A kind of method for carrying out real-time intrusion detection to civil small-scale unmanned plane using channel state information |
| CN108387940A (en) * | 2018-01-24 | 2018-08-10 | 西北工业大学 | A kind of detecting indoor article movement detection method based on Wi-Fi wireless awares |
| CN109961100A (en) * | 2019-03-26 | 2019-07-02 | 北京邮电大学 | WIFI indoor occupant intrusion detection method based on Principal Component Analysis |
| CN110012496A (en) * | 2019-04-25 | 2019-07-12 | 江苏创通电子股份有限公司 | A kind of intrusion detection method and device based on WiFi |
| CN110855382A (en) * | 2019-12-02 | 2020-02-28 | 北京邮电大学 | Method and device for detecting human body activity data based on CSI (channel State information), and electronic equipment |
| CN111246589A (en) * | 2018-11-28 | 2020-06-05 | 华为技术有限公司 | Random access signal sending method and device |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107968689B (en) * | 2017-12-06 | 2020-07-31 | 北京邮电大学 | Perception identification method and device based on wireless communication signals |
-
2020
- 2020-06-11 CN CN202010533116.7A patent/CN111741471B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108199757A (en) * | 2018-01-15 | 2018-06-22 | 浙江大学 | A kind of method for carrying out real-time intrusion detection to civil small-scale unmanned plane using channel state information |
| CN108387940A (en) * | 2018-01-24 | 2018-08-10 | 西北工业大学 | A kind of detecting indoor article movement detection method based on Wi-Fi wireless awares |
| CN111246589A (en) * | 2018-11-28 | 2020-06-05 | 华为技术有限公司 | Random access signal sending method and device |
| CN109961100A (en) * | 2019-03-26 | 2019-07-02 | 北京邮电大学 | WIFI indoor occupant intrusion detection method based on Principal Component Analysis |
| CN110012496A (en) * | 2019-04-25 | 2019-07-12 | 江苏创通电子股份有限公司 | A kind of intrusion detection method and device based on WiFi |
| CN110855382A (en) * | 2019-12-02 | 2020-02-28 | 北京邮电大学 | Method and device for detecting human body activity data based on CSI (channel State information), and electronic equipment |
Non-Patent Citations (4)
| Title |
|---|
| A Multi-View Discriminant Learning Approach for Indoor Localization Using Amplitude and Phase Features of CSI;Tahsina Farah Sanam;IEEE Access;全文 * |
| 一种基于CSI的非合作式人体行为识别方法;李晓薇;余江;常俊;杨锦朋;冉亚鑫;;计算机科学(第12期);全文 * |
| 一种面向新型入侵的获取和分类方法;王飞;周鹏程;王雷;徐本连;;计算机科学(第S3期);全文 * |
| 基于Wi-Fi网络的单AP室内定位方案的研究;孙广文等;中国优秀硕士学位论文全文数据库;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111741471A (en) | 2020-10-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111741471B (en) | Intrusion detection method and device based on CSI and computer storage medium | |
| CN107968689B (en) | Perception identification method and device based on wireless communication signals | |
| CN109379379B (en) | Network intrusion detection method based on improved convolutional neural network | |
| Arshad et al. | Leveraging transfer learning in multiple human activity recognition using WiFi signal | |
| CN106846729B (en) | Tumble detection method and system based on convolutional neural network | |
| CN108197581B (en) | Unmanned aerial vehicle signal identification detection method based on improved AC-WGANs | |
| US20220172378A1 (en) | Image processing apparatus, image processing method and non-transitory computer readable medium | |
| US9218540B2 (en) | Apparatus and computer readable medium for signal classification using spectrogram and templates | |
| CN109462610A (en) | A kind of network inbreak detection method based on Active Learning and transfer learning | |
| CN111698258A (en) | WiFi-based environmental intrusion detection method and system | |
| CN111753686A (en) | CSI-based people number identification method, device, equipment and computer storage medium | |
| Zhang et al. | A dynamic hand gesture recognition algorithm based on CSI and YOLOv3 | |
| CN112949487A (en) | WiFi-based ship cab personnel number detection method and system | |
| Tseng et al. | A graph Fourier transform based method for missing temperature data detection | |
| CN117272143A (en) | Power transmission line fault identification method and device based on gram angle field and residual error network | |
| Gulghane et al. | A survey on intrusion detection system using machine learning algorithms | |
| Arya et al. | Ensemble filter-based feature selection model for cyber attack detection in industrial Internet of Things | |
| CN110830504A (en) | Network intrusion behavior detection method and system | |
| Zhao et al. | Detection of unmanned aerial vehicle signal based on Gaussian mixture model | |
| CN108199757B (en) | A method of it is invaded using channel state information detection consumer level unmanned plane | |
| CN111476059A (en) | Target detection method and device, computer equipment and storage medium | |
| CN110046639A (en) | A kind of Hyperspectral imaging noise label detection method based on super-pixel weight density | |
| CN103077376B (en) | Method for distinguishing is known again based on the human body image in video image | |
| EP4254241A1 (en) | Method and device for image-based malware detection, and artificial intelligence-based endpoint detection and response system using same | |
| CN113452700A (en) | Method, device, equipment and storage medium for processing safety information |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |