CN108038419A - Wi-Fi-based indoor personnel passive detection method - Google Patents
Wi-Fi-based indoor personnel passive detection method Download PDFInfo
- Publication number
- CN108038419A CN108038419A CN201711137357.4A CN201711137357A CN108038419A CN 108038419 A CN108038419 A CN 108038419A CN 201711137357 A CN201711137357 A CN 201711137357A CN 108038419 A CN108038419 A CN 108038419A
- Authority
- CN
- China
- Prior art keywords
- mrow
- mtr
- mtd
- msubsup
- msub
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/02—Preprocessing
- G06F2218/04—Denoising
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B7/00—Radio transmission systems, i.e. using radiation field
- H04B7/02—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas
- H04B7/04—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas
- H04B7/08—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the receiving station
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W84/00—Network topologies
- H04W84/18—Self-organising networks, e.g. ad-hoc networks or sensor networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/08—Feature extraction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/12—Classification; Matching
Abstract
The invention discloses a Wi-Fi-based indoor personnel passive detection method. Establishing reasonable indoor measurement point distribution in an off-line stage, completing data acquisition in the off-line stage, and selecting data of a part of measurement points for off-line training; after the original data are exported, carrying out noise reduction treatment on the original data to generate a probability density function PDF and determine a threshold value; using a sliding window and calculating a sliding window correlation coefficient matrix of each receiving antenna; calculating a correlation coefficient matrix TT of data received by a receiving antenna in an online state, drawing a PDF image of the matrix TT, comparing the maximum point of the image with a threshold value, judging whether a test state is a static state, and determining the specific state of a target in a current scene. The method is based on the fine-grained physical layer information CSI, the correlation among the subcarriers is calculated on the frequency domain to generate the fingerprint, the influence of the wireless signals on the time domain by environmental factors is avoided, the occurrence of false alarm can be reduced through the threshold value determined by multiple times of training and the voting mode, and the detection accuracy is high.
Description
Technical field
The invention belongs to wireless sensor network field, is related to a kind of indoor occupant passive detection method based on Wi-Fi,
Accuracy and the robustness deficiency had a great influence with solving existing passive type personnel detection method by environmental change, detected is asked
Topic, while in the presence of target is detected, the active state current to target determines whether.
Background technology
With the rapid development of network technology, wireless signal spreads all over each corner of life substantially.Nowadays, people carry out
The time of outdoor activity and work gradually decreases, and most activity all carries out indoors, such as works, has a meal, does shopping, entertains
Etc..Passive type indoor occupant detection based on channel condition information (Channel State Information, CSI) feature
Method can be applied to smart home, medical monitoring, safety detection etc..
Indoors in environment, wireless signal reaches receiving terminal by reflection, diffraction and scattering etc. by mulitpath, such as schemes
Shown in 1.The wireless signal in different paths has different delays, decay and phase place change, and what receiving terminal received is that these are multiple
The signal that miscellaneous wireless signal fusion forms.Since interior is there are during personnel, either static target or moving target can
The wireless signal propagated in space is had an impact, therefore, by comparing under scene unmanned state and there are nothing under dbjective state
The characteristic difference of line signal, it can be determined that go out in scene and whether there is personnel.
Variation by analyzing wireless signal under different scenes can complete indoor occupant detection, wherein Fingerprint Model
Method is current mainstream technology, and this method, which exists, consumes small, the gain of parameter relatively advantage such as simple, precision height.Based on fingerprint mould
The method of type is made of offline and online two parts:Under off-line state, unmanned environment is adopted respectively and there are wireless communication during target
Number, extract characteristic fingerprint;Under presence, gather wireless signal, the characteristic fingerprint under extraction characteristic fingerprint and off-line state into
Row compares, judge under scene someone or nobody.But existing detection method is typically that characteristic fingerprint is extracted in time domain, by
Environment has a great influence, and there is limitation, and accuracy rate is relatively low, these problems all need to be further improved.
, can be with since received signal intensity indicates (Received Signal Strength Indication, RSSI)
Easily obtained by wireless technology and Cellular Networks etc., therefore traditional wireless aware means use RSSI.But RSSI is matchmaker
The energy response of body intervention key-course (Media Access Control, MAC), is easily influenced, very not by multipath effect
Stablize, be only applicable to the scene that environment is simple and required precision is not high.Specifically, in a typical indoor environment, nothing
, it is necessary to by reflection, diffraction and scattering etc. during the entire process of line signal is propagated, receiving terminal is reached eventually through multiple paths
's.Wireless signal from different paths has different delays, decay and phase place change, some receiver is in some time
Point is received is merged the signal formed by these complicated wireless signals.RSSI is folded as the signal from mulitpath
It is value added, it is highly unstable.An even static link without human interference, can also produce ripple on different time points
It is dynamic.This is because although the fluctuation on a certain paths is very small, may cause after the fluctuation superposition of all links
Last superposition value produces violent change.
Channel condition information is physical layer characteristic, and in wireless communication field, CSI refers to the channel attribute of communication link, it
Describe the decay factor in communication process of the signal between signal projector and receiver, including scattering, environmental attenuation, away from
From information such as decay, three in the mulitpath that CSI is propagated indoors are as shown in Fig. 2, Fig. 3 is that three paths are opposite in Fig. 2
The signal waveform answered.CSI generally comprises instantaneous CSI and statistical CSI, wherein, instantaneous CSI can be with channel condition known conditions
It is considered as the impulse response of digital filter, by instantaneous CSI signal projector and receiver can be allowed to adapt to transmission signal
Impulse response, so as to optimize spatial reuse and reduce data transmission fault rate;Statistical CSI is one of channel long-term observation result
Statistical value, the distribution situation, average channel gain and space phase of the signal attenuation factor are will be seen that by this statistical information
The information such as closing property.CSI is compared to more fine granularity and stabilization for RSSI, but accurate obtain needs special equipment, this becomes
Obstruction during initial research.However as orthogonal frequency division multiplexi (Orthogonal Frequency Division
Multiplexing, OFDM) widely use, researcher can by change firmware method will obtain CSI in different sons
Sampled version channel frequency response (Channel Frequency Response, CFR) in carrier frequency, therefore nearest grind
Study carefully all to turn to and utilize more fine-grained CSI information.
The existing detection method based on CSI is all that wireless signal is handled in time domain and extracts characteristic fingerprint, but when
Wireless signal can be effected by environmental factors on domain, there is limitation, and accuracy rate is relatively low.
The content of the invention
It is an object of the invention to solve to detect scene unmanned state in existing method and there are dynamic object shape
The fingerprint extracted in state, and time domain is influenced by environmental factor and multipath effect, and there are larger limitation.The present invention
Based on fine granularity physical layer information CSI, the validity feature fingerprint on frequency domain is extracted, unmanned scene is trained when offline, there are static
Target and there are characteristic fingerprint during moving target scene, for every reception antenna, is extracted by analyzing under different conditions
The signal characteristic gone out, determines to distinguish the threshold value of nobody and someone, and draws there are static target and there are general during moving target
Rate distribution histogram;On-line stage, by calculating the signal characteristic of sampled signal, it is first determined whether there is detection target,
In the presence of detecting target, current probability distribution histogram is further calculated, passes through the static state and dynamic good with precondition
Histogram is compared, and judges to detect the current state of target.After the testing result of every reception antenna is obtained, throwing is utilized
Ticket scheme determines final testing result.
To achieve the above object, the technical solution adopted by the present invention is a kind of indoor occupant passive detection based on Wi-Fi
Method, specifically comprises the steps of:
Step 1:Off-line phase, establishes rational indoor measurement point distribution, some by being divided into after scene removing barrier
A region, the data in measurement point when gathering static target first;Secondly walked at random in measured zone during collection dynamic object
Dynamic data;Data when finally gathering in scene without measured target, complete the data acquisition of off-line phase, therefrom choose one
Divide the data of measurement point to be used for off-line training, reduce training cost;
Step 2:Export the initial data of measurement;
Step 3:Noise reduction process is carried out to initial data, abnormal point is removed by wave filter, while disappear using medium filtering
Except the noise influenced in CSI communication processes be subject to multipath effect and environment in be mingled in initial data;
Step 4:Calculate scene nobody when and there are the related coefficient C on frequency domain during targetsta、With
Generating probability density function PDF, according to the intersection point threshold value δ of PDF images;
Step 5:Continue to analyze characteristic fingerprint of the scene there are static target and there are dynamic object using sliding window,
Using sliding window and calculate the sliding window correlation matrix C of every reception antennastan_aveAnd Cdyn_ave, draw
Cstan_aveAnd Cdyn_aveHistogram frequency distribution diagram;
Step 6:On-line stage, gathers the online data of one group of any duration, calculates every reception antenna under presence
The correlation matrix TT of data is received, draws the PDF images of matrix TT, the size of movement images peak and threshold value δ, judges
Whether test mode is static state, if the test value on no less than two antennas draws current shape with threshold value δ result of the comparison
State is divided there are target, then to continue to calculate the correlation matrix TT ' in sliding window on frequency domain on every antenna in scene
Probability distribution histograms of the measurement data TT ' under three antenna conditions is not drawn, by it with there are static target and existing dynamic
Probability distribution histogram during state target is compared, and there are the particular state of target under the high as current scene of similarity.
Preferably, above-mentioned wave filter uses hampel wave filters.
Further, above-mentioned steps four specifically include:
Step 4.1:Under scene unmanned state, the related coefficient between subcarrier two-by-two is calculated, every reception antenna receives
Data generation static fingerprint CstaFor:
Wherein,For the related coefficient under scene unmanned state between the i-th subcarriers and i+1 subcarriers;
Step 4.2:Have under static target state, equally calculate related coefficient, the data life that every reception antenna receives
Into the fingerprint there are static targetFor:
Wherein,For there are the related coefficient under static target state between the i-th subcarriers and i+1 subcarriers,
Cstan_lFor the fingerprint vector in a measurement point, l is measurement position point;
Step 4.3:There are under dynamic object state, related coefficient is calculated, the data generation that every reception antenna receives
The fingerprint there are dynamic objectFor:
Wherein,For there are the related coefficient under dynamic object state between the i-th subcarriers and i+1 subcarriers,
Cdyn_lFor the fingerprint vector in a measurement point, l is measurement position point;
Step 4.4:By Csta、WithThe generating probability density function PDF in a figure, it is possible to find nobody and
Differentiation during someone is obvious, the differentiation unobvious there are static target and there are dynamic object, therefore compares PDF image peaks
Numerical value, threshold value δ marks the line of demarcation of scene someone and unmanned state.
Further, above-mentioned steps five include:
Step 5.1:There are the characteristic fingerprint C for the data generation that during static target, every reception antenna receivesstan_ave
For:
Wherein,The average value of characteristic fingerprint between subcarrier under each sliding window;
Step 5.2:There are the characteristic fingerprint C for the data generation that during dynamic object, every reception antenna receivesdyn_ave
For:
Wherein,The average value of characteristic fingerprint between subcarrier under each sliding window.
Preferably, sliding window described in above-mentioned steps five is 5 seconds.
Equally, preferably, sliding window described in above-mentioned steps six is also configured as 5 seconds.
Compared with prior art, beneficial effects of the present invention:
(1) equipment is unrelated, easy to implement.
Passive type indoor occupant detection method based on CSI characteristic fingerprints is device-independent, it, which only needs one, to send out
Send wireless signal general commercial router and equipped with 5300 network interface cards of Intel and the computer of CSI information can be exported can be complete
Into work, it is not necessary to increase extra wireless device, it is not required that target carries relevant sensor device, therefore cost it is low and
Easily implement.
(2) CSI possesses abundant signal characteristic, and tool has great advantage.
Relative to the single fingerprint characteristics of RSS, CSI possesses more rich finger print information, such as frequency decay, phase
And energy intensity etc., these features can reflect signal characteristic during target invasion, and be perceived more in time domain and frequency domain
Trickle environmental information.
(3) CSI has stronger stability
Under identical environment, for RSS, the integral structure characteristic of CSI is stablized relatively, has and keeps phase under static environment
To stablizing and to the characteristic of target motion sensitive, being more suitable for complicated indoor positioning environment.Meanwhile CSI also can be certain
Multipath transmisstion is portrayed in degree.
(4) Detection accuracy is high, can determine whether out the particular state of target
Indoor occupant detection method based on fingerprint is customized according to scene, and exclusive feature is trained according to the scene of utilization
Fingerprint.Effective CSI features are extracted in this programme, the correlation generation fingerprint between subcarrier is calculated on frequency domain, is avoided well
Wireless signal is effected by environmental factors in time domain, and mistake can be reduced by the definite threshold value of repeatedly training and ballot mode
The appearance of report, therefore Detection accuracy is high.In addition, this programme is not only to detect scene whether there is target, additionally it is possible to judges target
Particular state, there is diversity.
Brief description of the drawings
Fig. 1 is the schematic diagram for passing multipath propagation.
Fig. 2 be CSI indoors in communication process three paths signal schematic representation.
Fig. 3 is the signal waveforms in three different paths.
Fig. 4 is the layout for testing scene.
Fig. 5 is the passive type indoor occupant detection method flow chart based on CSI characteristic fingerprints under off-line state.
Fig. 6 is the passive type indoor occupant detection method flow chart based on CSI characteristic fingerprints under presence.
Embodiment
In conjunction with attached drawing, the present invention will be further described in detail.
The present invention is a kind of passive type indoor occupant detection method based on CSI characteristic fingerprints, in a stable channel
In, channel information can be by being in Modeling In Frequency Domain:
Y=Hx+n (13)
Wherein y and x is the vector of signal receiving end and transmitting terminal respectively, and H is channel information matrix, n for Gaussian noise to
Amount.By formula, we can obtain the CSI calculation formula of all subcarriers:
CSI on per subcarriers can be expressed as:
H=| H | ejsin(∠H) (15)
Wherein | H | and ∠ H are respectively the amplitude and phase of each subcarrier.
The present invention uses a wireless access points (Wireless in the indoor scene of long 8.6m wide 5.7m
Access Point, AP) i.e. wireless router, an and computer equipped with 3 antenna Intel, 5300 network interface cards is as monitoring point
(Monitoring Point, MP), obtains CSI data, experiment porch is based on " Linux802.11n CSI by changing firmware
The wireless aware system platform based on Wi-Fi that the integrated installation tool TNS-CSI Tool of Tool " are built, implements step
It is as follows:
Step 1:Rational indoor measurement point distribution is established, completes the data acquisition of off-line phase.
The present invention is based on CSI characteristic fingerprints, and the method for Fingerprint Model is divided into offline and online two stages.Off-line phase
Work and be:
(1) ground removed in scene outside desk, chair etc. plant oneself is divided into the region of 32 1m*1m, often
The central point in a region is set to measurement point.AP and MP are placed in scene both sides and on same straight line.As shown in Figure 4.
(2) measured target is still in measurement point, and there are 32 groups of data during static target for collection.
(3) the measured target random walk in 32 regions successively, there are 32 groups of data during moving target for collection.
(4) without measured target in scene, data when collection static state is without target.
Cost when in the present invention, in order to reduce off-line training, choose measurement point 1,3,5,7,9,11,13,15,18,
20th, 22,24,25,27,29, the CSI numerical value that totally 15 points receive is received as Offline training data, all 32 measurement points
To data be used to detect the accuracy of proposition method of the present invention.
Step 2:Initial data is exported, in the present invention, the CSI that every antenna collects is the matrix H (T=50* of 30*T
T, rate of giving out a contract for a project are 50packets/s, and t is the time) set, it is shown that from transmitting antenna to reception antenna different sub-carrier
Channel gain.Every antenna is in the CSI matrix Hs that a measurement point receives:
Wherein hiRepresent the CSI vectors of the i-th subcarriers:
Step 3:In wireless communications, can be influenced in the communication process of CSI be subject to multipath effect and environment, original number
According to inevitably mixing a degree of noise.Here abnormal point is removed using hampel wave filters:
H '=hampel (H, 50*k) (18)
Wherein, H is the initial CSI that measurement point is tested, and k is constant, and H' is the CSI matrixes eliminated after abnormal point.
After eliminating abnormal point, noise is eliminated using medium filtering:
M is constant,For the CSI data after noise reduction.
Step 4:Calculate scene nobody when and there are the related coefficient on frequency domain during target, generating probability density function
(Probability Density Function, PDF), according to the intersection point threshold value of PDF images.Arrived used in stepSub-carrier vector in the CSI matrixes received for the i-th subcarriers.
Step 4.1:Under scene unmanned state, the related coefficient between subcarrier two-by-two is calculated, every reception antenna receives
Data generation static fingerprint CstaFor:
Wherein,For the related coefficient under scene unmanned state between the i-th subcarriers and i+1 subcarriers.
Step 4.2:Have under static target state, equally calculate related coefficient, the data life that every reception antenna receives
Into the fingerprint there are static targetFor:
Wherein,For there are the related coefficient under static target state between the i-th subcarriers and i+1 subcarriers,
Cstan_lFor the fingerprint vector in a measurement point, l is measurement position point.
Step 4.3:There are under dynamic object state, related coefficient is calculated, the data generation that every reception antenna receives
The fingerprint there are dynamic objectFor:
Wherein,For there are the related coefficient under dynamic object state between the i-th subcarriers and i+1 subcarriers,
Cdyn_lFor the fingerprint vector in a measurement point, l is measurement position point.
Step 4.4 is by Csta、WithThe generating probability density function PDF in a figure, it is possible to find nobody and have
Differentiation during people is obvious, the differentiation unobvious there are static target and there are dynamic object, therefore compares PDF image peaks
Numerical value, threshold value δ mark the line of demarcation of scene someone and unmanned state.
Step:5:Using sliding window, by drawing scene there are static target and there are feature during dynamic object to refer to
The probability distribution histogram of line, and the probability distribution histogram of characteristic fingerprint that the data processing with being gathered under on-line stage goes out is made
Compare, determine whether the active state of indoor objects.In this step, 5 seconds are chosen as sliding window and calculates m related coefficient
Value,For j-th strip subcarrier reception to CSI matrixes in sub-carrier vector.
Step 5.1:There are the characteristic fingerprint C for the data generation that during static target, every reception antenna receivesstan_ave
For:
Wherein,The average value of characteristic fingerprint between subcarrier under each sliding window.
Step 5.2:There are the characteristic fingerprint C for the data generation that during dynamic object, every reception antenna receivesdyn_ave
For:
Wherein,The average value of characteristic fingerprint between subcarrier under each sliding window.
Step 6:The online data of one group of any duration is gathered, every reception antenna under presence is calculated and receives data
Correlation matrix TT:
The PDF images of matrix TT are drawn, the size of movement images peak and threshold value δ, judges whether test mode is quiet
State state, if the test value on no less than two antennas draws current state for there are mesh in scene with threshold value δ result of the comparison
Mark, then continue to calculate on every antenna the correlation matrix TT ' on using 5 seconds as sliding window time-frequency domain:
Probability distribution histograms of the correlation matrix TT ' under three antenna conditions under sliding window is drawn respectively, will
Its with there are static target and there are probability distribution histogram during dynamic object to be compared, similarity is high and to work as front court
There are the particular state of target under scape.
In combination with substantial amounts of experimental data, emulation experiment, the validity of verification method are carried out using Matlab.Offline rank
The idiographic flow of section is as shown in figure 5, the idiographic flow of on-line stage is as shown in Figure 6.
Claims (6)
1. a kind of indoor occupant passive detection method based on Wi-Fi, it is characterised in that comprise the steps of:
Step 1:Off-line phase, establishes rational indoor measurement point distribution, several areas is divided into after scene is removed barrier
Domain, the data in measurement point when gathering static target first;Secondly in measured zone random walk during collection dynamic object
Data;Data when finally gathering in scene without measured target, complete the data acquisition of off-line phase, therefrom choose a part and survey
The data of amount point are used for off-line training, reduce training cost;
Step 2:Export the initial data of measurement;
Step 3:Noise reduction process is carried out to initial data, abnormal point is removed by wave filter, while eliminate using medium filtering
In CSI communication processes the noise in be mingled in initial data is influenced be subject to multipath effect and environment;
Step 4:Calculate scene nobody when and there are the related coefficient C on frequency domain during targetsta、WithGeneration
Probability density function PDF, according to the intersection point threshold value δ of PDF images;
Step 5:Continue to analyze characteristic fingerprint of the scene there are static target and there are dynamic object using sliding window, use
Sliding window and the sliding window correlation matrix C for calculating every reception antennastan_aveAnd Cdyn_ave, draw Cstan_aveWith
Cdyn_aveHistogram frequency distribution diagram;
Step 6:On-line stage, gathers the online data of one group of any duration, calculates every reception antenna under presence and receives
The correlation matrix TT of data, draws the PDF images of matrix TT, the size of movement images peak and threshold value δ, judges to test
Whether state is static state, if the test value on no less than two antennas show that current state is with threshold value δ result of the comparison
There are target in scene, then continue to calculate the correlation matrix TT ' in sliding window on frequency domain on every antenna, draw respectively
Go out probability distribution histograms of the measurement data TT ' under three antenna conditions, by its with there are static target and there are dynamic mesh
The probability distribution histogram of timestamp is compared, and there are the particular state of target under the high as current scene of similarity.
2. the indoor occupant passive detection method according to claim 1 based on Wi-Fi, it is characterised in that the wave filter
For hampel wave filters.
3. the indoor occupant passive detection method according to claim 1 based on Wi-Fi, it is characterised in that step 4 is specific
Including:
Step 4.1:Under scene unmanned state, the related coefficient between subcarrier two-by-two, the number that every reception antenna receives are calculated
According to the static fingerprint C of generationstaFor:
<mrow>
<msub>
<mi>C</mi>
<mrow>
<mi>s</mi>
<mi>t</mi>
<mi>a</mi>
</mrow>
</msub>
<mo>=</mo>
<mfenced open = "[" close = "]">
<mtable>
<mtr>
<mtd>
<msubsup>
<mi>C</mi>
<mrow>
<mi>s</mi>
<mi>t</mi>
<mi>a</mi>
</mrow>
<mn>1</mn>
</msubsup>
</mtd>
</mtr>
<mtr>
<mtd>
<msubsup>
<mi>C</mi>
<mrow>
<mi>s</mi>
<mi>t</mi>
<mi>a</mi>
</mrow>
<mn>2</mn>
</msubsup>
</mtd>
</mtr>
<mtr>
<mtd>
<mo>.</mo>
</mtd>
</mtr>
<mtr>
<mtd>
<mo>.</mo>
</mtd>
</mtr>
<mtr>
<mtd>
<mo>.</mo>
</mtd>
</mtr>
<mtr>
<mtd>
<msubsup>
<mi>C</mi>
<mrow>
<mi>s</mi>
<mi>t</mi>
<mi>a</mi>
</mrow>
<mi>i</mi>
</msubsup>
</mtd>
</mtr>
<mtr>
<mtd>
<mo>.</mo>
</mtd>
</mtr>
<mtr>
<mtd>
<mo>.</mo>
</mtd>
</mtr>
<mtr>
<mtd>
<mo>.</mo>
</mtd>
</mtr>
<mtr>
<mtd>
<msubsup>
<mi>C</mi>
<mrow>
<mi>s</mi>
<mi>t</mi>
<mi>a</mi>
</mrow>
<mrow>
<mi>N</mi>
<mo>-</mo>
<mn>1</mn>
</mrow>
</msubsup>
</mtd>
</mtr>
</mtable>
</mfenced>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>1</mn>
<mo>)</mo>
</mrow>
</mrow>
<mrow>
<msubsup>
<mi>C</mi>
<mrow>
<mi>s</mi>
<mi>t</mi>
<mi>a</mi>
</mrow>
<mi>i</mi>
</msubsup>
<mo>=</mo>
<mi>c</mi>
<mi>o</mi>
<mi>r</mi>
<mi>r</mi>
<mrow>
<mo>(</mo>
<msub>
<mover>
<mi>h</mi>
<mo>~</mo>
</mover>
<mi>i</mi>
</msub>
<mo>,</mo>
<msub>
<mover>
<mi>h</mi>
<mo>~</mo>
</mover>
<mrow>
<mi>i</mi>
<mo>+</mo>
<mn>1</mn>
</mrow>
</msub>
<mo>)</mo>
</mrow>
<mo>,</mo>
<mi>i</mi>
<mo>&Element;</mo>
<mo>&lsqb;</mo>
<mn>1</mn>
<mo>,</mo>
<mi>N</mi>
<mo>-</mo>
<mn>1</mn>
<mo>&rsqb;</mo>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>2</mn>
<mo>)</mo>
</mrow>
</mrow>
Wherein,For the related coefficient under scene unmanned state between the i-th subcarriers and i+1 subcarriers;
Step 4.2:Have under static target state, equally calculate related coefficient, the data generation that every reception antenna receives
There are the fingerprint of static targetFor:
<mrow>
<msub>
<mover>
<mi>C</mi>
<mo>~</mo>
</mover>
<mrow>
<mi>s</mi>
<mi>tan</mi>
<mo>_</mo>
<mi>l</mi>
</mrow>
</msub>
<mo>=</mo>
<mfrac>
<mn>1</mn>
<mi>l</mi>
</mfrac>
<mo>&Sigma;</mo>
<mrow>
<mo>(</mo>
<msub>
<mi>C</mi>
<mrow>
<mi>s</mi>
<mi>tan</mi>
<mo>_</mo>
<mi>l</mi>
</mrow>
</msub>
<mo>)</mo>
</mrow>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>3</mn>
<mo>)</mo>
</mrow>
</mrow>
<mrow>
<msub>
<mi>C</mi>
<mrow>
<mi>s</mi>
<mi>tan</mi>
<mo>_</mo>
<mi>l</mi>
</mrow>
</msub>
<mo>=</mo>
<mfenced open = "[" close = "]">
<mtable>
<mtr>
<mtd>
<msubsup>
<mi>C</mi>
<mrow>
<mi>s</mi>
<mi>tan</mi>
<mo>_</mo>
<mi>l</mi>
</mrow>
<mn>1</mn>
</msubsup>
</mtd>
</mtr>
<mtr>
<mtd>
<msubsup>
<mi>C</mi>
<mrow>
<mi>s</mi>
<mi>tan</mi>
<mo>_</mo>
<mi>l</mi>
</mrow>
<mn>2</mn>
</msubsup>
</mtd>
</mtr>
<mtr>
<mtd>
<mo>.</mo>
</mtd>
</mtr>
<mtr>
<mtd>
<mo>.</mo>
</mtd>
</mtr>
<mtr>
<mtd>
<mo>.</mo>
</mtd>
</mtr>
<mtr>
<mtd>
<msubsup>
<mi>C</mi>
<mrow>
<mi>s</mi>
<mi>tan</mi>
<mo>_</mo>
<mi>l</mi>
</mrow>
<mi>i</mi>
</msubsup>
</mtd>
</mtr>
<mtr>
<mtd>
<mo>.</mo>
</mtd>
</mtr>
<mtr>
<mtd>
<mo>.</mo>
</mtd>
</mtr>
<mtr>
<mtd>
<mo>.</mo>
</mtd>
</mtr>
<mtr>
<mtd>
<msubsup>
<mi>C</mi>
<mrow>
<mi>s</mi>
<mi>tan</mi>
<mo>_</mo>
<mi>l</mi>
</mrow>
<mrow>
<mi>N</mi>
<mo>-</mo>
<mn>1</mn>
</mrow>
</msubsup>
</mtd>
</mtr>
</mtable>
</mfenced>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>4</mn>
<mo>)</mo>
</mrow>
</mrow>
<mrow>
<msubsup>
<mi>C</mi>
<mrow>
<mi>s</mi>
<mi>tan</mi>
<mo>_</mo>
<mi>l</mi>
</mrow>
<mi>i</mi>
</msubsup>
<mo>=</mo>
<mi>c</mi>
<mi>o</mi>
<mi>r</mi>
<mi>r</mi>
<mrow>
<mo>(</mo>
<msub>
<mover>
<mi>h</mi>
<mo>~</mo>
</mover>
<mi>i</mi>
</msub>
<mo>,</mo>
<msub>
<mover>
<mi>h</mi>
<mo>~</mo>
</mover>
<mrow>
<mi>i</mi>
<mo>+</mo>
<mn>1</mn>
</mrow>
</msub>
<mo>)</mo>
</mrow>
<mo>,</mo>
<mi>i</mi>
<mo>&Element;</mo>
<mo>&lsqb;</mo>
<mn>1</mn>
<mo>,</mo>
<mi>N</mi>
<mo>-</mo>
<mn>1</mn>
<mo>&rsqb;</mo>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>5</mn>
<mo>)</mo>
</mrow>
</mrow>
Wherein,For there are the related coefficient under static target state between the i-th subcarriers and i+1 subcarriers,
Cstan_lFor the fingerprint vector in a measurement point, l is measurement position point;
Step 4.3:There are under dynamic object state, related coefficient is calculated, the data generation that every reception antenna receives is deposited
In the fingerprint of dynamic objectFor:
<mrow>
<msub>
<mover>
<mi>C</mi>
<mo>~</mo>
</mover>
<mrow>
<mi>d</mi>
<mi>y</mi>
<mi>n</mi>
<mo>_</mo>
<mi>l</mi>
</mrow>
</msub>
<mo>=</mo>
<mfrac>
<mn>1</mn>
<mi>l</mi>
</mfrac>
<mo>&Sigma;</mo>
<mrow>
<mo>(</mo>
<msub>
<mi>C</mi>
<mrow>
<mi>d</mi>
<mi>y</mi>
<mi>n</mi>
<mo>_</mo>
<mi>l</mi>
</mrow>
</msub>
<mo>)</mo>
</mrow>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>6</mn>
<mo>)</mo>
</mrow>
</mrow>
<mrow>
<msub>
<mi>C</mi>
<mrow>
<mi>d</mi>
<mi>y</mi>
<mi>n</mi>
<mo>_</mo>
<mi>l</mi>
</mrow>
</msub>
<mo>=</mo>
<mfenced open = "[" close = "]">
<mtable>
<mtr>
<mtd>
<msubsup>
<mi>C</mi>
<mrow>
<mi>d</mi>
<mi>y</mi>
<mi>n</mi>
<mo>_</mo>
<mi>l</mi>
</mrow>
<mn>1</mn>
</msubsup>
</mtd>
</mtr>
<mtr>
<mtd>
<msubsup>
<mi>C</mi>
<mrow>
<mi>d</mi>
<mi>y</mi>
<mi>n</mi>
<mo>_</mo>
<mi>l</mi>
</mrow>
<mn>2</mn>
</msubsup>
</mtd>
</mtr>
<mtr>
<mtd>
<mo>.</mo>
</mtd>
</mtr>
<mtr>
<mtd>
<mo>.</mo>
</mtd>
</mtr>
<mtr>
<mtd>
<mo>.</mo>
</mtd>
</mtr>
<mtr>
<mtd>
<msubsup>
<mi>C</mi>
<mrow>
<mi>d</mi>
<mi>y</mi>
<mi>n</mi>
<mo>_</mo>
<mi>l</mi>
</mrow>
<mi>i</mi>
</msubsup>
</mtd>
</mtr>
<mtr>
<mtd>
<mo>.</mo>
</mtd>
</mtr>
<mtr>
<mtd>
<mo>.</mo>
</mtd>
</mtr>
<mtr>
<mtd>
<mo>.</mo>
</mtd>
</mtr>
<mtr>
<mtd>
<msubsup>
<mi>C</mi>
<mrow>
<mi>d</mi>
<mi>y</mi>
<mi>n</mi>
<mo>_</mo>
<mi>l</mi>
</mrow>
<mrow>
<mi>N</mi>
<mo>-</mo>
<mn>1</mn>
</mrow>
</msubsup>
</mtd>
</mtr>
</mtable>
</mfenced>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>7</mn>
<mo>)</mo>
</mrow>
</mrow>
<mrow>
<msubsup>
<mi>C</mi>
<mrow>
<mi>d</mi>
<mi>y</mi>
<mi>n</mi>
<mo>_</mo>
<mi>l</mi>
</mrow>
<mi>i</mi>
</msubsup>
<mo>=</mo>
<mi>c</mi>
<mi>o</mi>
<mi>r</mi>
<mi>r</mi>
<mrow>
<mo>(</mo>
<msub>
<mover>
<mi>h</mi>
<mo>~</mo>
</mover>
<mi>i</mi>
</msub>
<mo>,</mo>
<msub>
<mover>
<mi>h</mi>
<mo>~</mo>
</mover>
<mrow>
<mi>i</mi>
<mo>+</mo>
<mn>1</mn>
</mrow>
</msub>
<mo>)</mo>
</mrow>
<mo>,</mo>
<mi>i</mi>
<mo>&Element;</mo>
<mo>&lsqb;</mo>
<mn>1</mn>
<mo>,</mo>
<mi>N</mi>
<mo>-</mo>
<mn>1</mn>
<mo>&rsqb;</mo>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>8</mn>
<mo>)</mo>
</mrow>
</mrow>
Wherein,For there are the related coefficient under dynamic object state between the i-th subcarriers and i+1 subcarriers, Cdyn_l
For the fingerprint vector in a measurement point, l is measurement position point;
Step 4.4:By Csta、WithThe generating probability density function PDF in a figure, it is possible to find nobody and someone
When differentiation it is obvious, differentiation unobvious there are static target and there are dynamic object, therefore compare the number of PDF image peaks
Value, threshold value δ mark the line of demarcation of scene someone and unmanned state.
4. the indoor occupant passive detection method according to claim 1 based on Wi-Fi, it is characterised in that step 5 bag
Include:
Step 5.1:There are the characteristic fingerprint C for the data generation that during static target, every reception antenna receivesstan_aveFor:
<mrow>
<msub>
<mi>C</mi>
<mrow>
<mi>s</mi>
<mi>tan</mi>
<mo>_</mo>
<mi>a</mi>
<mi>v</mi>
<mi>e</mi>
</mrow>
</msub>
<mo>=</mo>
<mfenced open = "[" close = "]">
<mtable>
<mtr>
<mtd>
<msubsup>
<mi>C</mi>
<mrow>
<mi>s</mi>
<mi>tan</mi>
<mo>_</mo>
<mi>a</mi>
<mi>v</mi>
<mi>e</mi>
</mrow>
<mn>1</mn>
</msubsup>
</mtd>
</mtr>
<mtr>
<mtd>
<msubsup>
<mi>C</mi>
<mrow>
<mi>s</mi>
<mi>tan</mi>
<mo>_</mo>
<mi>a</mi>
<mi>v</mi>
<mi>e</mi>
</mrow>
<mn>2</mn>
</msubsup>
</mtd>
</mtr>
<mtr>
<mtd>
<mo>.</mo>
</mtd>
</mtr>
<mtr>
<mtd>
<mo>.</mo>
</mtd>
</mtr>
<mtr>
<mtd>
<mo>.</mo>
</mtd>
</mtr>
<mtr>
<mtd>
<msubsup>
<mi>C</mi>
<mrow>
<mi>s</mi>
<mi>tan</mi>
<mo>_</mo>
<mi>a</mi>
<mi>v</mi>
<mi>e</mi>
</mrow>
<mi>i</mi>
</msubsup>
</mtd>
</mtr>
<mtr>
<mtd>
<mo>.</mo>
</mtd>
</mtr>
<mtr>
<mtd>
<mo>.</mo>
</mtd>
</mtr>
<mtr>
<mtd>
<mo>.</mo>
</mtd>
</mtr>
<mtr>
<mtd>
<msubsup>
<mi>C</mi>
<mrow>
<mi>s</mi>
<mi>tan</mi>
<mo>_</mo>
<mi>a</mi>
<mi>v</mi>
<mi>e</mi>
</mrow>
<mi>m</mi>
</msubsup>
</mtd>
</mtr>
</mtable>
</mfenced>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>9</mn>
<mo>)</mo>
</mrow>
</mrow>
<mrow>
<msubsup>
<mi>C</mi>
<mrow>
<mi>s</mi>
<mi>tan</mi>
<mo>_</mo>
<mi>a</mi>
<mi>v</mi>
<mi>e</mi>
</mrow>
<mi>i</mi>
</msubsup>
<mo>=</mo>
<mfrac>
<mn>1</mn>
<mrow>
<mi>N</mi>
<mo>-</mo>
<mn>1</mn>
</mrow>
</mfrac>
<mo>&Sigma;</mo>
<mi>c</mi>
<mi>o</mi>
<mi>r</mi>
<mi>r</mi>
<mrow>
<mo>(</mo>
<msub>
<mover>
<mi>h</mi>
<mo>~</mo>
</mover>
<mi>j</mi>
</msub>
<mo>,</mo>
<msub>
<mover>
<mi>h</mi>
<mo>~</mo>
</mover>
<mrow>
<mi>j</mi>
<mo>+</mo>
<mn>1</mn>
</mrow>
</msub>
<mo>)</mo>
</mrow>
<mo>,</mo>
<mi>j</mi>
<mo>&Element;</mo>
<mo>&lsqb;</mo>
<mn>1</mn>
<mo>,</mo>
<mi>N</mi>
<mo>-</mo>
<mn>1</mn>
<mo>&rsqb;</mo>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>10</mn>
<mo>)</mo>
</mrow>
</mrow>
Wherein,The average value of characteristic fingerprint between subcarrier under each sliding window;
Step 5.2:There are the characteristic fingerprint C for the data generation that during dynamic object, every reception antenna receivesdyn_aveFor:
<mrow>
<msub>
<mi>C</mi>
<mrow>
<mi>d</mi>
<mi>y</mi>
<mi>n</mi>
<mo>_</mo>
<mi>a</mi>
<mi>v</mi>
<mi>e</mi>
</mrow>
</msub>
<mo>=</mo>
<mfenced open = "[" close = "]">
<mtable>
<mtr>
<mtd>
<msubsup>
<mi>C</mi>
<mrow>
<mi>d</mi>
<mi>y</mi>
<mi>n</mi>
<mo>_</mo>
<mi>a</mi>
<mi>v</mi>
<mi>e</mi>
</mrow>
<mn>1</mn>
</msubsup>
</mtd>
</mtr>
<mtr>
<mtd>
<msubsup>
<mi>C</mi>
<mrow>
<mi>d</mi>
<mi>y</mi>
<mi>n</mi>
<mo>_</mo>
<mi>a</mi>
<mi>v</mi>
<mi>e</mi>
</mrow>
<mn>2</mn>
</msubsup>
</mtd>
</mtr>
<mtr>
<mtd>
<mo>.</mo>
</mtd>
</mtr>
<mtr>
<mtd>
<mo>.</mo>
</mtd>
</mtr>
<mtr>
<mtd>
<mo>.</mo>
</mtd>
</mtr>
<mtr>
<mtd>
<msubsup>
<mi>C</mi>
<mrow>
<mi>d</mi>
<mi>y</mi>
<mi>n</mi>
<mo>_</mo>
<mi>a</mi>
<mi>v</mi>
<mi>e</mi>
</mrow>
<mi>i</mi>
</msubsup>
</mtd>
</mtr>
<mtr>
<mtd>
<mo>.</mo>
</mtd>
</mtr>
<mtr>
<mtd>
<mo>.</mo>
</mtd>
</mtr>
<mtr>
<mtd>
<mo>.</mo>
</mtd>
</mtr>
<mtr>
<mtd>
<msubsup>
<mi>C</mi>
<mrow>
<mi>d</mi>
<mi>y</mi>
<mi>n</mi>
<mo>_</mo>
<mi>a</mi>
<mi>v</mi>
<mi>e</mi>
</mrow>
<mi>m</mi>
</msubsup>
</mtd>
</mtr>
</mtable>
</mfenced>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>11</mn>
<mo>)</mo>
</mrow>
</mrow>
<mrow>
<msubsup>
<mi>C</mi>
<mrow>
<mi>d</mi>
<mi>y</mi>
<mi>n</mi>
<mo>_</mo>
<mi>a</mi>
<mi>v</mi>
<mi>e</mi>
</mrow>
<mi>i</mi>
</msubsup>
<mo>=</mo>
<mfrac>
<mn>1</mn>
<mrow>
<mi>N</mi>
<mo>-</mo>
<mn>1</mn>
</mrow>
</mfrac>
<mo>&Sigma;</mo>
<mi>c</mi>
<mi>o</mi>
<mi>r</mi>
<mi>r</mi>
<mrow>
<mo>(</mo>
<msub>
<mover>
<mi>h</mi>
<mo>~</mo>
</mover>
<mi>j</mi>
</msub>
<mo>,</mo>
<msub>
<mover>
<mi>h</mi>
<mo>~</mo>
</mover>
<mrow>
<mi>j</mi>
<mo>+</mo>
<mn>1</mn>
</mrow>
</msub>
<mo>)</mo>
</mrow>
<mo>,</mo>
<mi>j</mi>
<mo>&Element;</mo>
<mo>&lsqb;</mo>
<mn>1</mn>
<mo>,</mo>
<mi>N</mi>
<mo>-</mo>
<mn>1</mn>
<mo>&rsqb;</mo>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>12</mn>
<mo>)</mo>
</mrow>
</mrow>
Wherein,The average value of characteristic fingerprint between subcarrier under each sliding window.
5. the indoor occupant passive detection method according to claim 1 based on Wi-Fi, it is characterised in that institute in step 5
Sliding window is stated as 5 seconds.
6. the indoor occupant passive detection method according to claim 1 based on Wi-Fi, it is characterised in that institute in step 6
Sliding window is stated as 5 seconds.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711137357.4A CN108038419B (en) | 2017-11-16 | 2017-11-16 | Wi-Fi-based indoor personnel passive detection method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711137357.4A CN108038419B (en) | 2017-11-16 | 2017-11-16 | Wi-Fi-based indoor personnel passive detection method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN108038419A true CN108038419A (en) | 2018-05-15 |
CN108038419B CN108038419B (en) | 2022-05-13 |
Family
ID=62092762
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201711137357.4A Active CN108038419B (en) | 2017-11-16 | 2017-11-16 | Wi-Fi-based indoor personnel passive detection method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108038419B (en) |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109151731A (en) * | 2018-06-14 | 2019-01-04 | 西北师范大学 | A kind of indoor orientation method based on CSI |
CN109784282A (en) * | 2019-01-18 | 2019-05-21 | 重庆邮电大学 | Passive type personnel motion detection and method for tracing based on signal coherence feature |
CN111781592A (en) * | 2020-06-12 | 2020-10-16 | 中国船舶重工集团公司第七二四研究所 | Rapid automatic starting method based on fine-grained characteristic analysis |
CN112034433A (en) * | 2020-07-09 | 2020-12-04 | 重庆邮电大学 | Through-wall passive moving target detection method based on interference signal reconstruction |
CN112163540A (en) * | 2020-10-09 | 2021-01-01 | 上海第二工业大学 | Gesture recognition method based on WiFi |
CN112869734A (en) * | 2021-01-11 | 2021-06-01 | 乐鑫信息科技(上海)股份有限公司 | Wi-Fi human body detection method and intelligent device |
CN113408691A (en) * | 2021-06-22 | 2021-09-17 | 哈尔滨工程大学 | Method for predicting through-wall passive population based on channel state information |
CN114422050A (en) * | 2021-12-13 | 2022-04-29 | 华南师范大学 | Method, system and platform for detecting subcarriers and visualizing subcarriers in real time based on WIFI |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102883360A (en) * | 2012-10-30 | 2013-01-16 | 无锡儒安科技有限公司 | Method and system for wirelessly omnidirectionally and passively detecting user indoors |
CN105873212A (en) * | 2016-05-16 | 2016-08-17 | 南京邮电大学 | Indoor-environment-person detection method based on channel state information |
CN106658590A (en) * | 2016-12-28 | 2017-05-10 | 南京航空航天大学 | Design and implementation of multi-person indoor environment state monitoring system based on WiFi channel state information |
CN106792808A (en) * | 2016-12-08 | 2017-05-31 | 南京邮电大学 | Los path recognition methods under a kind of indoor environment based on channel condition information |
-
2017
- 2017-11-16 CN CN201711137357.4A patent/CN108038419B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102883360A (en) * | 2012-10-30 | 2013-01-16 | 无锡儒安科技有限公司 | Method and system for wirelessly omnidirectionally and passively detecting user indoors |
CN105873212A (en) * | 2016-05-16 | 2016-08-17 | 南京邮电大学 | Indoor-environment-person detection method based on channel state information |
CN106792808A (en) * | 2016-12-08 | 2017-05-31 | 南京邮电大学 | Los path recognition methods under a kind of indoor environment based on channel condition information |
CN106658590A (en) * | 2016-12-28 | 2017-05-10 | 南京航空航天大学 | Design and implementation of multi-person indoor environment state monitoring system based on WiFi channel state information |
Non-Patent Citations (1)
Title |
---|
朱海等: "基于信道状态信息的WiFi环境感知技术", 《南京邮电大学学报(自然科学版)》 * |
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109151731A (en) * | 2018-06-14 | 2019-01-04 | 西北师范大学 | A kind of indoor orientation method based on CSI |
CN109784282A (en) * | 2019-01-18 | 2019-05-21 | 重庆邮电大学 | Passive type personnel motion detection and method for tracing based on signal coherence feature |
CN111781592A (en) * | 2020-06-12 | 2020-10-16 | 中国船舶重工集团公司第七二四研究所 | Rapid automatic starting method based on fine-grained characteristic analysis |
CN111781592B (en) * | 2020-06-12 | 2023-12-22 | 中国船舶集团有限公司第七二四研究所 | Rapid automatic starting method based on fine granularity characteristic analysis |
CN112034433A (en) * | 2020-07-09 | 2020-12-04 | 重庆邮电大学 | Through-wall passive moving target detection method based on interference signal reconstruction |
CN112034433B (en) * | 2020-07-09 | 2024-01-12 | 深圳市领冠检测技术有限公司 | Through-wall passive moving target detection method based on interference signal reconstruction |
CN112163540A (en) * | 2020-10-09 | 2021-01-01 | 上海第二工业大学 | Gesture recognition method based on WiFi |
CN112163540B (en) * | 2020-10-09 | 2024-01-19 | 上海第二工业大学 | Gesture recognition method based on WiFi |
CN112869734A (en) * | 2021-01-11 | 2021-06-01 | 乐鑫信息科技(上海)股份有限公司 | Wi-Fi human body detection method and intelligent device |
CN113408691A (en) * | 2021-06-22 | 2021-09-17 | 哈尔滨工程大学 | Method for predicting through-wall passive population based on channel state information |
CN114422050A (en) * | 2021-12-13 | 2022-04-29 | 华南师范大学 | Method, system and platform for detecting subcarriers and visualizing subcarriers in real time based on WIFI |
Also Published As
Publication number | Publication date |
---|---|
CN108038419B (en) | 2022-05-13 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108038419A (en) | Wi-Fi-based indoor personnel passive detection method | |
Choi et al. | Deep learning based NLOS identification with commodity WLAN devices | |
Zhang et al. | An indoor positioning method based on CSI by using features optimization mechanism with LSTM | |
CN109672485B (en) | Indoor personnel real-time invasion and movement speed detection method based on channel state information | |
CN103209478B (en) | Based on the indoor orientation method of classification thresholds and signal strength signal intensity weight | |
CN104502894B (en) | Method for passive detection of moving objects based on physical layer information | |
CN102883360B (en) | A kind of method and system of the passive user's detection of indoor wireless omnidirectional | |
CN106792808A (en) | Los path recognition methods under a kind of indoor environment based on channel condition information | |
CN107480699A (en) | A kind of intrusion detection method based on channel condition information and SVMs | |
CN106658590A (en) | Design and implementation of multi-person indoor environment state monitoring system based on WiFi channel state information | |
CN103916954B (en) | Probabilistic Localization Methods and positioner based on WLAN | |
KR100948412B1 (en) | Method and system for estimating the location using a Received Signal Strength Indication | |
CN106802404A (en) | A kind of human body indoor positioning automatic testing method and system | |
CN109698724A (en) | Intrusion detection method, device, equipment and storage medium | |
CN114222202B (en) | Environment self-adaptive activity detection method and system based on WiFi CSI | |
Samadh et al. | Indoor localization based on channel state information | |
CN110730473A (en) | Signal feature extraction method for WiFi activity recognition | |
CN106793085A (en) | Fingerprint positioning method based on normality assumption inspection | |
Tsai et al. | Refined autoencoder-based CSI hidden feature extraction for indoor spot localization | |
CN111901028A (en) | Human body behavior identification method based on CSI (channel State information) on multiple antennas | |
CN105866732A (en) | Improved MK model and WKNN algorithm combined mixed indoor positioning method | |
Fan et al. | Single-site indoor fingerprint localization based on MIMO-CSI | |
CN113453180A (en) | Intelligent detection method and system for human body tumble and information data processing terminal | |
CN107911863A (en) | A kind of method that malice AP positions are determined based on simple gesture | |
Yang | WiLocus: CSI based human tracking system in indoor environment |
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 |