CN108416974A - Autoalarm based on radio channel status information and method - Google Patents
Autoalarm based on radio channel status information and method Download PDFInfo
- Publication number
- CN108416974A CN108416974A CN201810063069.7A CN201810063069A CN108416974A CN 108416974 A CN108416974 A CN 108416974A CN 201810063069 A CN201810063069 A CN 201810063069A CN 108416974 A CN108416974 A CN 108416974A
- Authority
- CN
- China
- Prior art keywords
- data
- personnel
- action
- state information
- channel state
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 28
- 230000009471 action Effects 0.000 claims abstract description 97
- 238000004891 communication Methods 0.000 claims abstract description 52
- 239000013598 vector Substances 0.000 claims description 27
- 239000011159 matrix material Substances 0.000 claims description 20
- 230000000875 corresponding effect Effects 0.000 claims description 19
- 238000005070 sampling Methods 0.000 claims description 16
- 238000013135 deep learning Methods 0.000 claims description 12
- 230000008859 change Effects 0.000 claims description 11
- 238000012545 processing Methods 0.000 claims description 9
- 238000011946 reduction process Methods 0.000 claims description 7
- 239000000284 extract Substances 0.000 claims description 4
- 230000007246 mechanism Effects 0.000 claims description 4
- 230000017105 transposition Effects 0.000 claims description 4
- 230000002159 abnormal effect Effects 0.000 description 4
- 230000008901 benefit Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 3
- 230000011218 segmentation Effects 0.000 description 3
- 230000006978 adaptation Effects 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000003672 processing method Methods 0.000 description 2
- 238000012546 transfer Methods 0.000 description 2
- 230000001154 acute effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 239000000969 carrier Substances 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 210000000056 organ Anatomy 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
- G08B21/02—Alarms for ensuring the safety of persons
- G08B21/0297—Robbery alarms, e.g. hold-up alarms, bag snatching alarms
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B17/00—Monitoring; Testing
- H04B17/30—Monitoring; Testing of propagation channels
Landscapes
- Physics & Mathematics (AREA)
- Electromagnetism (AREA)
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Business, Economics & Management (AREA)
- Emergency Management (AREA)
- General Physics & Mathematics (AREA)
- Alarm Systems (AREA)
Abstract
The invention discloses a kind of autoalarm and method based on radio channel status information, the radio channel status information are launch terminal and to receive the channel state information of the communication subcarrier between terminal in cordless communication network, which includes：Alarm module, alarm module for obtaining channel state information, and according to channel state information in cordless communication network overlay area personnel's gesture or personnel's action be identified, and warning message is generated according to recognition result；Communication module, communication module are alarmed for receiving warning message according to warning message.Autoalarm according to the present invention can easily and effectively realize automatic alarm, and cost is relatively low.
Description
Technical field
The present invention relates to wireless communication technology field, more particularly to a kind of automatic report based on radio channel status information
Alarm device and a kind of automatic alarm method based on radio channel status information.
Background technology
Home security is always problem of people's attention, with the development of science and technology, requirement of the people for security protection is also to get over
Come higher, the alarm system of mainstream is all based on touch-tone initiative alarming and sensor-based passive alarm.Pass through button
The initiative alarming of formula has some limitations.At the large stretch of region of monitoring, sensor-based passive alarm needs
It monitors and arranges a certain number of sensors in region, and the increase of the quantity with sensor, the expenditure of cost can be increased, and
And sensor-based passive alarm system is also required to people and is alarmed by button to put through communication apparatus, works as the freedom of person
In the case of confined, it can have some limitations.
Invention content
The present invention is directed to solve the technical problems such as of high cost, the alarm inconvenience of alarm system at least to a certain extent, it is
This, an object of the present invention is to provide a kind of autoalarms based on radio channel status information, can conveniently have
Automatic alarm is realized on effect ground, and cost is relatively low.
Second object of the present invention is to propose a kind of automatic alarm method based on radio channel status information.
In order to achieve the above objectives, first aspect present invention embodiment propose based on the automatic of radio channel status information
Warning device, wherein the radio channel status information is logical between launch terminal and reception terminal in cordless communication network
Believe that the channel state information of subcarrier, the autoalarm include：Alarm module, the alarm module is for obtaining institute
Channel state information is stated, and according to the channel state information to personnel's gesture in the cordless communication network overlay area
Or personnel's action is identified, and warning message is generated according to recognition result；Communication module, the communication module is for connecing
The warning message is received, and is alarmed according to the warning message.
Autoalarm according to the ... of the embodiment of the present invention based on radio channel status information, alarm module can obtain
Channel state information, and according to channel state information in cordless communication network overlay area personnel's gesture or personnel act
It is identified, and warning message is generated according to recognition result, communication module can alarm according to warning message, sharp as a result,
Channel state information is used as the physical quantity of action recognition, has many advantages, such as stable, reliable, precision height, without identification blind area, closes
Existing wireless telecom equipment is utilized in reason, at low cost, is easy to universal, and carries any active equipment without human body, into one
Step reduces cost, and in general, the autoalarm of the embodiment of the present invention can easily and effectively realize automatic alarm,
And cost is relatively low.
In addition, also according to the autoalarm based on radio channel status information of the above embodiment of the present invention proposition
There can be following additional technical characteristic：
Further, the alarm module includes：Data processing unit, for the channel state information into line number
Data preprocess；Continuous action recognition unit, for by continuous action recognizer to carrying out obtained letter after data prediction
Channel state information data carry out continuous action identification, to obtain effective action data segment；Cutting unit is acted, for described
Effective action data segment is split to obtain multiple single movement data；Recognition unit, for according to the single movement number
Personnel's gesture or personnel action are identified according to deep learning algorithm.
Specifically, the channel state information is sampled to obtain by the reception terminal with preset sample frequency, every time
It samples obtained channel state information data and is expressed as a matrix, the data processing unit is used for：It is filtered by Hampel
Device, Butterworth filters carry out noise reduction process to the channel state information data；Pass through the method for weighted moving average pair
The channel state information data after noise reduction process are carried out to be reconstructed.
Specifically, the continuous action recognition unit is used for：By the sliding window of preset window size with default step-length
Interception carries out the channel state information data obtained after data prediction, to obtain multiple subdivision matrixes；It will each segmentation
Matrix is multiplied with the transposition of itself, to obtain corresponding correlation matrix；Calculate each correlation matrix characteristic value and feature to
Amount, and judge the characteristic value of multiple correlation matrixes and the situation of change of feature vector；It is determined whether according to the situation of change
Specific action substantially acts, with the determination effective action data segment.
Specifically, the action cutting unit is used for：Estimate each acted in the effective action data segment
Initial point and end point obtain estimating set：Wherein,I-th respectively estimated
The starting point and end point of a action, andConstitute a data pair；Set personal distance parameter Tb, and pass through
The personal distance parameter, to being extended, obtains new set to each data estimated in set：It is multiple independent to be obtained according to the new set
Action data.
Specifically, be stored with sample data set in the recognition unit, wherein will indicate in advance each personnel's gesture and
The sample data of personnel's action learns in deep learning algorithm, and extracts corresponding sampling feature vectors, with according to the sample
Eigen vector establishes the sample data set, and the recognition unit is used for：Using deep learning algorithm to each described independent
Action data is trained, to obtain the feature vector of corresponding action to be identified；By the feature vector of the action to be identified with
The sampling feature vectors that the sample data is concentrated are compared, to identify that personnel's gesture or the personnel act.
Further, the alarm module further includes：Initiative alarming unit, the initiative alarming unit are used for described
When recognition unit identifies personnel's gesture, alarm number is gone out according to personnel's gesture identification, and by the alarm number
It is sent to the communication module；Passive alarm unit, the passive alarm unit are used to identify in the recognition unit described
When personnel act, the type of personnel's action is further judged, and endanger in the type judgement acted according to the personnel
Corresponding alert information is generated when dangerous, and the alert information is sent to the communication module.
Further, the communication module carries out dialing alarm when receiving the alarm number, and receiving
It alarms to the corresponding mechanism that receives a crime report when stating alert information.
In order to achieve the above objectives, second aspect of the present invention embodiment proposes a kind of based on radio channel status information
Automatic alarm method, wherein the radio channel status information is between launch terminal in cordless communication network and reception terminal
Communication subcarrier channel state information, the automatic alarm method includes：Obtain the channel state information；According to institute
State channel state information in the cordless communication network overlay area personnel's gesture or personnel's action be identified, and root
Warning message is generated according to recognition result；It is alarmed according to the warning message.
Automatic alarm method according to the ... of the embodiment of the present invention based on radio channel status information, can obtain channel status
Information, and according to channel state information in cordless communication network overlay area personnel's gesture or personnel action be identified,
And warning message is generated according to recognition result, and can be alarmed according to warning message, channel state information is utilized as a result,
As the physical quantity of action recognition, has many advantages, such as stable, reliable, precision height, without identification blind area, be rationally utilized existing
Wireless telecom equipment, it is at low cost, it is easy to universal, and any active equipment is carried without human body, further reduces costs, always
For body, the automatic alarm method of the embodiment of the present invention can easily and effectively realize automatic alarm.
In addition, also according to the automatic alarm method based on radio channel status information of the above embodiment of the present invention proposition
There can be following additional technical characteristic：
Further, according to the channel state information to personnel's gesture in the cordless communication network overlay area
Or personnel's action is identified, and specifically includes：Data prediction is carried out to the channel state information；Known by continuous action
Other algorithm carries out continuous action identification to the channel state information data obtained after data prediction, effectively to be moved
Make data segment；The effective action data segment is split to obtain multiple single movement data；According to the single movement
Data and deep learning algorithm identify personnel's gesture or personnel action.
Description of the drawings
Fig. 1 is the structural schematic diagram according to the cordless communication network of one embodiment of the invention；
Fig. 2 is the box according to the autoalarm based on radio channel status information of one embodiment of the invention
Schematic diagram；
Fig. 3 is the structure according to the autoalarm based on radio channel status information of one embodiment of the invention
Schematic diagram；
Fig. 4 is the flow according to the automatic alarm method based on radio channel status information of one embodiment of the invention
Figure.
Specific implementation mode
The embodiment of the present invention is described below in detail, examples of the embodiments are shown in the accompanying drawings, wherein from beginning to end
Same or similar label indicates same or similar element or element with the same or similar functions.Below with reference to
The embodiment of attached drawing description is exemplary, it is intended to for explaining the present invention, and is not considered as limiting the invention.
Below in conjunction with the accompanying drawings come describe the embodiment of the present invention autoalarm based on radio channel status information and
Method.
The radio channel status information of the embodiment of the present invention is between launch terminal in cordless communication network and reception terminal
Communication subcarrier channel state information.Wherein, cordless communication network can be WIFI (Wireless Fidelity, Yi Zhongwu
Line LAN), launch terminal may include router, receives terminal and may include configured with wireless network card and connect the router
Terminal device, such as mobile phone, tablet computer, laptop etc..
As shown in Figure 1, launch terminal and the channel of reception terminal consist of a plurality of sub-carriers, passed when human body is in signal
Defeated space and when making action, can influence the propagation of signal.In the ofdm system of one embodiment of the invention
In, it is made of 30 sub- carrier waves per a pair of launch terminal and the channel for receiving terminal, when human body does one near reception terminal
When a little actions, the variation of the phase value or amplitude of the channel state information CSI on 30 subcarriers, and this can be caused
Changing influences faint action to be also significant.It therefore, in an embodiment of the present invention, can be according to the letter of subcarrier
Channel state information carries out action recognition.
In one embodiment of the invention, channel state information CSI can be carried out by reception terminal with preset sample frequency
Sampling obtains, and samples obtained channel state information data every time and is represented by a matrix.Specifically, receiving terminal can be with
The sample frequency of 2000 packets/second samples channel state information data, and sampling every time obtains matrixBy embedded ofdm system, by the channel status
Information data preserves, and action recognition and alarm etc. next will be carried out based on the channel state information data.
Fig. 2 is the box according to the autoalarm based on radio channel status information of one embodiment of the invention
Schematic diagram.
As shown in Fig. 2, the autoalarm based on radio channel status information of the embodiment of the present invention, including alarm
Module 10 and communication module 20.
Wherein, alarm module 10 is used to obtain channel state information, and according to channel state information to cordless communication network
Personnel's gesture or personnel's action in overlay area are identified, and generate warning message according to recognition result；Communication module
20 for receiving warning message, and is alarmed according to the warning message.
Alarm module 10, which may be disposed at, to be received in terminal, can also be independently disposed to receive except terminal, when being independently disposed to
When receiving except terminal, alarm module 10 can be carried out wireless communication with terminal is received to receive the channel status that it is collected
Information.
Further, as shown in Fig. 2, alarm module 10 may include data processing unit 11, continuous action recognition unit
12, cutting unit 13 and recognition unit 14 are acted.Wherein, data processing unit 11 is used to carry out data to channel state information
Pretreatment；Continuous action recognition unit 12 is used for through continuous action recognizer to carrying out obtained letter after data prediction
Channel state information data carry out continuous action identification, to obtain effective action data segment；Cutting unit 13 is acted to be used for effective
Action data section is split to obtain multiple single movement data；Recognition unit 14 is used for according to single movement data and depth
Spend learning algorithm identification personnel's gesture or personnel's action.
In one embodiment of the invention, data processing unit 11 be specifically used for by Hampel filters,
Butterworth filters carry out noise reduction process to channel state information data, and by the method for weighted moving average to dropping
Making an uproar treated, channel state information data are reconstructed.Due to signal in transfer process the change of transimission power and biography
The influence of the factors such as the adaptation selection of defeated rate, can inevitably cause to sample has some prominent in obtained channel state information data
The value of change.The median that μ is sampled data can be defined according to the principle of Hampel filters, it is in sampled data to define σ
The absolute deviation of digit defines the normal range (NR) that [μ-γ × σ, μ+γ × σ] is the channel state information data that sampling obtains,
Except this range is exceptional value.In addition, since personnel's operating frequency is typically very low, then channel state information number
It is also a kind of exception according to the very high point of middle frequency, can one cutoff frequency be set according to the principle of Butterworth filters, it is right
Data are further processed, and eliminate the very high abnormal data of frequency.By above two data processing method, although having eliminated
Most of abnormal data, but still have noise that can influence the subsequent processing to data in data, so mobile flat using weighting
The principle of equal method, passes throughWeight
Structure channel state information data, to keep channel state information data variation smoother.
In one embodiment of the invention, continuous action recognition unit 12 is specifically used for through preset window size
Sliding window intercepts the channel state information data for carrying out being obtained after data prediction with default step-length, to obtain multiple segmentations
Matrix, and each subdivision matrix is multiplied with the transposition of itself, to obtain corresponding correlation matrix, and calculate each related
The characteristic value and feature vector of matrix, and judge the characteristic value of multiple correlation matrixes and the situation of change of feature vector, then root
Specific action is determined whether according to situation of change or is substantially acted, to determine effective action data segment.Wherein, preset window is big
Small to be wrapped for 500, default step-length can be 400 packets.
In one embodiment of the invention, action cutting unit 13 is specifically used for estimating every in effective action data segment
The starting point and end point of a action obtain estimating set：Wherein,Respectively
The starting point and end point for i-th of the action estimated, andConstitute a data pair.Concurrently set personal distance parameter
Tb, and new set is obtained to estimating each data in set to being extended by personal distance parameter：Determine the starting point each acted and end
Point, so as to obtain multiple single movement data according to the new set.
In one embodiment of the invention, it is stored with sample data set in recognition unit 14, wherein can be in advance by table
Show that each personnel's gesture and the sample data of personnel's action learn in deep learning algorithm, and extracts corresponding sample characteristics
Vector, to establish the sample data set according to sampling feature vectors.Recognition unit 14 is specifically used for utilizing deep learning algorithm pair
Each single movement data are trained, to obtain the feature vector of corresponding action to be identified, and by the feature of action to be identified
Vector is compared with the sampling feature vectors that sample data is concentrated, to identify that personnel's gesture or personnel act.
In one embodiment of the invention, as shown in Fig. 2, alarm module 10 may also include 15 He of initiative alarming unit
Passive alarm unit 16.
Wherein, initiative alarming unit 15 is used for when recognition unit identifies personnel's gesture, is gone out according to personnel's gesture identification
Alarm number, and alarm number is sent to communication module 20.Communication module 20 can dial when receiving alarm number
Alarm.
Passive alarm unit 16 is used to, when recognition unit identifies personnel's action, further judge the class of personnel's action
Type, and corresponding alert information is generated when the type judgement acted according to personnel is caused danger, and alert information is sent
To communication module 20.Communication module 20 can alarm when receiving alert information to the corresponding mechanism that receives a crime report.
It in one particular embodiment of the present invention, can be in family or public place application autoalarm.It can be same
Two positions on one horizontal line put launch terminal and receive terminal respectively, as shown in figure 3, the launch terminal can be common
Wireless router, the reception terminal can by equipped with 5300 wireless network cards of Intel smart machine and reception antenna form.
As shown in figure 3, when the personnel of communication coverage area continuously, repeatedly make defined digital sign language, intelligence
Energy equipment can identify Connected digits according to channel state information, analyze the operator's field, site field and sequence of number
Then row field connects communication module to identify alarm number, realize the initiative alarming of gesture dialing.
When smart machine monitors to have in communication coverage area significantly physical attacks action, such as strikes, falls down, is acute
It is strong to run, it can determine whether out generation merit, communication module can be reported a case to the security authorities merit by internet to local public security organ at this time,
Realize passive alarm.Smart machine can also act according to the physical attacks of identification and determine merit property, and fire such as occurs, needs
First aid, generation house robbery etc., communication module can be alarmed according to merit property to corresponding responsible person and responsible organization.
It should be noted that it can be different or same intelligent terminal to receive terminal and autoalarm.For same
The case where one intelligent terminal, for example, when it is mobile phone to receive terminal, alarm module 10 may be disposed in mobile phone, lead to
News module 20 may also set up in mobile phone, i.e., the mobile phone can either be according to channel state information to the cordless communication network area of coverage
Personnel's gesture or personnel's action in domain are identified, and generate warning message according to recognition result, and can be according to alarm signal
Breath is alarmed.
Autoalarm according to the ... of the embodiment of the present invention based on radio channel status information, alarm module can obtain
Channel state information, and according to channel state information in cordless communication network overlay area personnel's gesture or personnel act
It is identified, and warning message is generated according to recognition result, communication module can alarm according to warning message, sharp as a result,
Channel state information is used as the physical quantity of action recognition, has many advantages, such as stable, reliable, precision height, without identification blind area, closes
Existing wireless telecom equipment is utilized in reason, at low cost, is easy to universal, and carries any active equipment without human body, further
Cost is reduced, in general, the autoalarm of the embodiment of the present invention can easily and effectively realize automatic alarm, and
Cost is relatively low.
Corresponding above-described embodiment, the present invention also propose a kind of automatic alarm method based on radio channel status information.
As shown in figure 4, the automatic alarm method based on radio channel status information of the embodiment of the present invention, including it is following
Step：
S1 obtains channel state information.
In one embodiment of the invention, channel state information CSI can be carried out by reception terminal with preset sample frequency
Sampling obtains, and samples obtained channel state information data every time and is represented by a matrix.Specifically, receiving terminal can be with
The sample frequency of 2000 packets/second samples channel state information data, and sampling every time obtains matrixBy embedded ofdm system, by the channel status
Information data preserves, and action recognition and alarm etc. next will be carried out based on the channel state information data.
S2, according to channel state information in cordless communication network overlay area personnel's gesture or personnel act carry out
Identification, and warning message is generated according to recognition result.
Specifically, data prediction can be carried out to channel state information, and by continuous action recognizer into line number
The channel state information data that are obtained after Data preprocess carry out continuous action identification, to obtain effective action data segment and right
Effective action data segment is split to obtain multiple single movement data, and is calculated according to single movement data and deep learning
Method identifies personnel's gesture or personnel's action.
It in one embodiment of the invention, can be by Hampel filters, Butterworth filters to channel shape
State information data carries out noise reduction process, and by the method for weighted moving average to the channel state information number after carrying out noise reduction process
According to being reconstructed.Due to signal in transfer process the factors such as adaptation selection of the change of transimission power and transmission rate
It influences, can inevitably cause to sample the value for thering are some to be mutated in obtained channel state information data.It can be according to Hampel filters
Principle, define μ be sampled data median, define σ be sampled data median absolute deviation, definition [μ-γ ×
σ, μ+γ × σ] it is the normal range (NR) for sampling obtained channel state information data, except this range is exceptional value.
In addition, since personnel's operating frequency is typically very low, then the very high point of frequency is also a kind of different in channel state information data
Often, one cutoff frequency can be set, data are further processed according to the principle of Butterworth filters, eliminates frequency
The very high abnormal data of rate.By above two data processing method, although having eliminated most of abnormal data, number
Still there is noise that can influence the subsequent processing to data in, so using the principle of the method for weighted moving average, passes throughReconstruct channel shape
State information data, to keep channel state information data variation smoother.
In one embodiment of the invention, continuous action recognition unit 12 is specifically used for through preset window size
Sliding window intercepts the channel state information data for carrying out being obtained after data prediction with default step-length, to obtain multiple segmentations
Matrix, and each subdivision matrix is multiplied with the transposition of itself, to obtain corresponding correlation matrix, and calculate each related
The characteristic value and feature vector of matrix, and judge the characteristic value of multiple correlation matrixes and the situation of change of feature vector, then root
Specific action is determined whether according to situation of change or is substantially acted, to determine effective action data segment.Wherein, preset window is big
Small to be wrapped for 500, default step-length can be 400 packets.
In one embodiment of the invention, the starting point and end point each acted in effective action data segment can be estimated,
It obtains estimating set：Wherein,The starting point and knot for i-th of the action respectively estimated
Spot, andConstitute a data pair.Concurrently set personal distance parameter Tb, and by personal distance parameter to estimating set
In each data to being extended, obtain new set：I.e.
The starting point and end point each acted is determined, so as to obtain multiple single movement data according to the new set.
In one embodiment of the invention, the sample data of each personnel's gesture and personnel's action will can be indicated in advance
Learn in deep learning algorithm, and extract corresponding sampling feature vectors, to establish sample data according to sampling feature vectors
Collection, and stored.Each single movement data can be trained using deep learning algorithm in identification, to obtain phase
Answer the feature vector of action to be identified, and the sampling feature vectors that the feature vector of action to be identified and sample data are concentrated
It is compared, to identify that personnel's gesture or personnel act.
S3 alarms according to warning message.
In one embodiment of the invention, can be by identifying that personnel's gesture realizes initiative alarming, it also can be by identifying people
Passive alarm is realized in member's action.Specifically, when recognition unit identifies personnel's gesture, report can be gone out according to personnel's gesture identification
Then warning signal code can carry out dialing alarm according to alarm number.When recognition unit identifies personnel's action, further judge
The type of personnel's action, and when the type judgement acted according to personnel is caused danger, reported to the corresponding mechanism that receives a crime report
It is alert.
Automatic alarm method according to the ... of the embodiment of the present invention based on radio channel status information, can obtain channel status
Information, and according to channel state information in cordless communication network overlay area personnel's gesture or personnel action be identified,
And warning message is generated according to recognition result, and can be alarmed according to warning message, channel state information is utilized as a result,
As the physical quantity of action recognition, has many advantages, such as stable, reliable, precision height, without identification blind area, be rationally utilized existing
Wireless telecom equipment, it is at low cost, it is easy to universal, and any active equipment is carried without human body, further reduces costs, always
For body, the automatic alarm method of the embodiment of the present invention can easily and effectively realize automatic alarm.
In the description of the present invention, it is to be understood that, term "center", " longitudinal direction ", " transverse direction ", " length ", " width
Degree ", " thickness ", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outside",
The orientation or positional relationship of the instructions such as " clockwise ", " counterclockwise ", " axial direction ", " radial direction ", " circumferential direction " is based on ... shown in the drawings
Orientation or positional relationship is merely for convenience of description of the present invention and simplification of the description, do not indicate or imply the indicated device or
Element must have a particular orientation, with specific azimuth configuration and operation, therefore be not considered as limiting the invention.
In addition, term " first ", " second " are used for description purposes only, it is not understood to indicate or imply relatively important
Property or implicitly indicate the quantity of indicated technical characteristic." first " is defined as a result, the feature of " second " can be expressed
Or implicitly include one or more this feature.In the description of the present invention, the meaning of " plurality " is two or two
More than, unless otherwise specifically defined.
In the present invention unless specifically defined or limited otherwise, term " installation ", " connected ", " connection ", " fixation "
Equal terms shall be understood in a broad sense, for example, it may be being fixedly connected, may be a detachable connection, or integral；It can be machine
Tool connects, and can also be electrical connection；It can be directly connected, can also can be indirectly connected through an intermediary two members
The interaction relationship of connection or two elements inside part.It for the ordinary skill in the art, can be according to tool
Body situation understands the concrete meaning of above-mentioned term in the present invention.
In the present invention unless specifically defined or limited otherwise, fisrt feature can be with "above" or "below" second feature
It is that the first and second features are in direct contact or the first and second features pass through intermediary mediate contact.Moreover, fisrt feature
Second feature " on ", " top " and " above " but fisrt feature be directly above or diagonally above the second feature, or only table
Show that fisrt feature level height is higher than second feature.Fisrt feature second feature " under ", " lower section " and " below " can be
Fisrt feature is directly under or diagonally below the second feature, or is merely representative of fisrt feature level height and is less than second feature.
In the description of this specification, reference term " one embodiment ", " some embodiments ", " example ", " specifically show
The description of example " or " some examples " etc. means specific features, structure, material or spy described in conjunction with this embodiment or example
Point is included at least one embodiment or example of the invention.In the present specification, schematic expression of the above terms are not
It must be directed to identical embodiment or example.Moreover, particular features, structures, materials, or characteristics described can be in office
It can be combined in any suitable manner in one or more embodiments or example.In addition, without conflicting with each other, this field
Technical staff can carry out the feature of different embodiments or examples described in this specification and different embodiments or examples
In conjunction with and combination.
Although the embodiments of the present invention has been shown and described above, it is to be understood that above-described embodiment is example
Property, it is not considered as limiting the invention, those skilled in the art within the scope of the invention can be to above-mentioned
Embodiment is changed, changes, replacing and modification.
Claims (10)
1. a kind of autoalarm based on radio channel status information, which is characterized in that the radio channel status information
For launch terminal in cordless communication network and receive the channel state information for communicating subcarrier between terminal, the automatic alarm
Device includes：
Alarm module, the alarm module are used to obtain the channel state information, and according to the channel state information to institute
It states the action of personnel's gesture or personnel in cordless communication network overlay area to be identified, and is generated and alarmed according to recognition result
Information；
Communication module, the communication module are alarmed for receiving the warning message according to the warning message.
2. the autoalarm according to claim 1 based on radio channel status information, which is characterized in that the report
Warning module includes：
Data processing unit, for carrying out data prediction to the channel state information；
Continuous action recognition unit, for by continuous action recognizer to carrying out obtained channel status after data prediction
Information data carries out continuous action identification, to obtain effective action data segment；
Cutting unit is acted, for being split to the effective action data segment to obtain multiple single movement data；
Recognition unit, for identifying personnel's gesture or the personnel according to the single movement data and deep learning algorithm
Action.
3. the autoalarm according to claim 2 based on radio channel status information, which is characterized in that the letter
Channel state information is sampled to obtain by the reception terminal with preset sample frequency, samples obtained channel state information every time
Data are expressed as a matrix, and the data processing unit is specifically used for：
Noise reduction process is carried out to the channel state information data by Hampel filters, Butterworth filters；
The channel state information data after carrying out noise reduction process are reconstructed by the method for weighted moving average.
4. the autoalarm according to claim 3 based on radio channel status information, which is characterized in that the company
Continuous action recognition unit is specifically used for：
The channel status obtain after data prediction is intercepted with default step-length by the sliding window of preset window size to believe
Data are ceased, to obtain multiple subdivision matrixes；
Each subdivision matrix is multiplied with the transposition of itself, to obtain corresponding correlation matrix；
It calculates the characteristic value and feature vector of each correlation matrix, and judges the characteristic value and feature vector of multiple correlation matrixes
Situation of change；
Specific action is determined whether according to the situation of change or is substantially acted, with the determination effective action data segment.
5. the autoalarm according to claim 4 based on radio channel status information, which is characterized in that described dynamic
It is specifically used for as cutting unit：
The starting point and end point each acted in the effective action data segment is estimated, obtains estimating set：Wherein,The starting point and end point for i-th of the action respectively estimated, andConstitute a data pair；
Set personal distance parameter Tb, and by the personal distance parameter to it is described estimate set in each data to carry out
Extension, obtains new set：With according to described new
Set obtain multiple single movement data.
6. the autoalarm according to claim 5 based on radio channel status information, which is characterized in that the knowledge
It is stored with sample data set in other unit, wherein will indicate the sample data of each personnel's gesture and personnel's action in depth in advance
Learn in degree learning algorithm, and extract corresponding sampling feature vectors, to establish the sample according to the sampling feature vectors
Data set, the recognition unit are specifically used for：
Each single movement data are trained using deep learning algorithm, to obtain the feature of corresponding action to be identified
Vector；
The feature vector of the action to be identified is compared with the sampling feature vectors that the sample data is concentrated, with identification
Personnel's gesture or personnel action.
7. the autoalarm based on radio channel status information according to any one of claim 2-6, feature
It is, the alarm module further includes：
Initiative alarming unit, the initiative alarming unit are used for when the recognition unit identifies personnel's gesture, according to
Personnel's gesture identification goes out alarm number, and the alarm number is sent to the communication module；
Passive alarm unit, the passive alarm unit are used for when the recognition unit identifies personnel's action, into one
Step judges the type of personnel's action, and generation is corresponding alert when the type judgement acted according to the personnel is caused danger
Feelings information, and the alert information is sent to the communication module.
8. the autoalarm according to claim 7 based on radio channel status information, which is characterized in that described logical
News module carries out dialing alarm when receiving the alarm number, and when receiving the alert information to receiving a crime report accordingly
Mechanism alarms.
9. a kind of automatic alarm method based on radio channel status information, which is characterized in that the radio channel status information
For launch terminal in cordless communication network and receive the channel state information for communicating subcarrier between terminal, the automatic alarm
Method includes：
Obtain the channel state information；
According to the channel state information in the cordless communication network overlay area personnel's gesture or personnel act carry out
Identification, and warning message is generated according to recognition result；
It is alarmed according to the warning message.
10. the automatic alarm method according to claim 9 based on radio channel status information, which is characterized in that according to
The channel state information in the cordless communication network overlay area personnel's gesture or personnel action be identified, specifically
Including：
Data prediction is carried out to the channel state information；
Continuous action is carried out to the channel state information data obtained after data prediction by continuous action recognizer
Identification, to obtain effective action data segment；
The effective action data segment is split to obtain multiple single movement data；
Personnel's gesture or personnel action are identified according to the single movement data and deep learning algorithm.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810063069.7A CN108416974B (en) | 2018-01-23 | 2018-01-23 | Automatic alarm device and method based on wireless channel state information |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810063069.7A CN108416974B (en) | 2018-01-23 | 2018-01-23 | Automatic alarm device and method based on wireless channel state information |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108416974A true CN108416974A (en) | 2018-08-17 |
| CN108416974B CN108416974B (en) | 2020-07-07 |
Family
ID=63126238
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810063069.7A Active CN108416974B (en) | 2018-01-23 | 2018-01-23 | Automatic alarm device and method based on wireless channel state information |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108416974B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109460716A (en) * | 2018-10-19 | 2019-03-12 | 大连理工大学 | A kind of sign language wireless-identification device and method |
| CN110706463A (en) * | 2019-06-11 | 2020-01-17 | 南京信息工程大学 | WIFI passive sensing method and system suitable for tumble monitoring |
| CN112596024A (en) * | 2020-12-04 | 2021-04-02 | 华中科技大学 | Motion identification method based on environment background wireless radio frequency signal |
| CN113359816A (en) * | 2021-05-21 | 2021-09-07 | 中国人民解放军陆军工程大学 | Unmanned aerial vehicle control method and system based on wireless gesture recognition |
| CN113643522A (en) * | 2021-08-31 | 2021-11-12 | 中国银行股份有限公司 | Alarm prediction method, device, equipment and storage medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104123007A (en) * | 2014-07-29 | 2014-10-29 | 电子科技大学 | Multidimensional weighted 3D recognition method for dynamic gestures |
| CN104615244A (en) * | 2015-01-23 | 2015-05-13 | 深圳大学 | Automatic gesture recognizing method and system |
| CN105761407A (en) * | 2016-01-06 | 2016-07-13 | 深圳大学 | Indoor fire detection and alarming method and system based on wireless network signal transmission |
| CN105915473A (en) * | 2016-05-26 | 2016-08-31 | 中南大学 | OFDM (Orthogonal Frequency Division Multiplexing) system parametric channel estimation and equalization method based on compressed sensing technology |
| CN106131958A (en) * | 2016-08-09 | 2016-11-16 | 电子科技大学 | A kind of based on channel condition information with the indoor Passive Location of support vector machine |
| CN107102729A (en) * | 2017-04-05 | 2017-08-29 | 河南师范大学 | A kind of PPT Demonstration Control Systems based on CSI gesture identifications |
-
2018
- 2018-01-23 CN CN201810063069.7A patent/CN108416974B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104123007A (en) * | 2014-07-29 | 2014-10-29 | 电子科技大学 | Multidimensional weighted 3D recognition method for dynamic gestures |
| CN104615244A (en) * | 2015-01-23 | 2015-05-13 | 深圳大学 | Automatic gesture recognizing method and system |
| CN105761407A (en) * | 2016-01-06 | 2016-07-13 | 深圳大学 | Indoor fire detection and alarming method and system based on wireless network signal transmission |
| CN105915473A (en) * | 2016-05-26 | 2016-08-31 | 中南大学 | OFDM (Orthogonal Frequency Division Multiplexing) system parametric channel estimation and equalization method based on compressed sensing technology |
| CN106131958A (en) * | 2016-08-09 | 2016-11-16 | 电子科技大学 | A kind of based on channel condition information with the indoor Passive Location of support vector machine |
| CN107102729A (en) * | 2017-04-05 | 2017-08-29 | 河南师范大学 | A kind of PPT Demonstration Control Systems based on CSI gesture identifications |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109460716A (en) * | 2018-10-19 | 2019-03-12 | 大连理工大学 | A kind of sign language wireless-identification device and method |
| CN110706463A (en) * | 2019-06-11 | 2020-01-17 | 南京信息工程大学 | WIFI passive sensing method and system suitable for tumble monitoring |
| CN112596024A (en) * | 2020-12-04 | 2021-04-02 | 华中科技大学 | Motion identification method based on environment background wireless radio frequency signal |
| CN113359816A (en) * | 2021-05-21 | 2021-09-07 | 中国人民解放军陆军工程大学 | Unmanned aerial vehicle control method and system based on wireless gesture recognition |
| CN113359816B (en) * | 2021-05-21 | 2022-07-15 | 中国人民解放军陆军工程大学 | Unmanned aerial vehicle control method and system based on wireless gesture recognition |
| CN113643522A (en) * | 2021-08-31 | 2021-11-12 | 中国银行股份有限公司 | Alarm prediction method, device, equipment and storage medium |
| CN113643522B (en) * | 2021-08-31 | 2023-06-06 | 中国银行股份有限公司 | Alarm prediction method, device, equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108416974B (en) | 2020-07-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108416974A (en) | Autoalarm based on radio channel status information and method | |
| US11212650B2 (en) | Device-free localization methods within smart indoor environments | |
| Choi et al. | People counting using IR-UWB radar sensor in a wide area | |
| CN107480699A (en) | A kind of intrusion detection method based on channel condition information and SVMs | |
| US10310079B1 (en) | Presence detection using wireless signals confirmed with ultrasound and/or audio | |
| Yang et al. | Fine-grained adaptive location-independent activity recognition using commodity WiFi | |
| Chen et al. | Taming the inconsistency of Wi-Fi fingerprints for device-free passive indoor localization | |
| Ponnusamy et al. | Primary user emulation attack mitigation using neural network | |
| CN110149604B (en) | Personnel detection method based on WIFI signal | |
| CN105405260B (en) | A kind of burglary-resisting system and theft preventing method based on wireless signal | |
| Liu et al. | Harvesting ambient RF for presence detection through deep learning | |
| CN114222202B (en) | Environment self-adaptive activity detection method and system based on WiFi CSI | |
| CN108038419A (en) | Wi-Fi-based indoor personnel passive detection method | |
| Zhu et al. | An environmental intrusion detection technology based on WiFi | |
| CN109255874B (en) | Method for detecting passing and number of people based on common commercial WiFi equipment | |
| US20240031774A1 (en) | Device-free localization methods within smart indoor environments | |
| Bahadori et al. | ReWiS: Reliable Wi-Fi sensing through few-shot multi-antenna multi-receiver CSI learning | |
| CN106899667A (en) | A kind of Internet of Things green house of vegetables intelligent security guard gateway and control method | |
| Yi et al. | LoRa Signal Monitoring System of Multi-Node Software Define Radio | |
| Hsieh et al. | CSI ratio with coloring-assisted learning for NLoS motionless human presence detection | |
| Verma et al. | On the performance of energy detection-based CR with SC diversity over IG channel | |
| US20230072968A1 (en) | System and Method for Sensing an Environment | |
| Cai et al. | CSI-based device-free indoor localization using convolutional neural networks | |
| CN113077600B (en) | Wi-Fi indoor security alarm system based on Fresnel zone | |
| Noguchi et al. | A CSI-based object detection scheme using interleaved subcarrier selection in wireless LAN systems with distributed antennas |
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 | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20240320 Address after: 221000 No. 6, Keji Road, Xuzhou Economic and Technological Development Zone, Xuzhou City, Jiangsu Province Patentee after: XCMG Hanyun Technology Co.,Ltd. Country or region after: Zhong Guo Address before: 221116 No. 1 Tongshan University Road, Xuzhou City, Jiangsu Province Patentee before: CHINA University OF MINING AND TECHNOLOGY Country or region before: Zhong Guo |
|
| TR01 | Transfer of patent right |