CN106618497A - Method for monitoring sleep in complicated environment based on channel state information - Google Patents
Method for monitoring sleep in complicated environment based on channel state information Download PDFInfo
- Publication number
- CN106618497A CN106618497A CN201611145695.8A CN201611145695A CN106618497A CN 106618497 A CN106618497 A CN 106618497A CN 201611145695 A CN201611145695 A CN 201611145695A CN 106618497 A CN106618497 A CN 106618497A
- Authority
- CN
- China
- Prior art keywords
- csi
- user
- phase
- information
- channel condition
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/48—Other medical applications
- A61B5/4806—Sleep evaluation
- A61B5/4815—Sleep quality
Abstract
The invention relates to a method for monitoring sleep in a complicated environment based on channel state information and belongs to the technical field of wireless network application. The method comprises the steps that Wi-Fi channel state information CSI in a target environment is firstly acquired through an existing Wi-Fi device, phase conversion is conducted on the CSI obtained after an outlier is removed to calibrate phase information, then the CSI without the outlier and CSI information obtained after phase calibration are used for judging whether an interference object moving around a user exists or not, interfered information is deleted if the moving object exists, accordingly accurate breathing information is obtained, the breathing frequency of the user is obtained according to the breathing information, and further the sleeping quality of the user is judged by analyzing the breathing frequency of the user. Compared with the prior art, the environment robustness and accuracy of sleep monitoring are greatly improved by adopting the method, the method does not depend on various sensors and wearable devices, and the method is low in cost, strong in anti-interference property and good in user experience and does not produce the privacy leakage problem.
Description
Technical field
The present invention relates to a kind of method of monitoring sleep, is supervised under more particularly to a kind of complex environment based on channel condition information
The method for surveying sleep, belongs to wireless network applied technical field, and in particular to channel state information of wireless network application process skill
Art field, for monitoring breathing state and sleep quality.
Background technology
Sleep, as process necessary to life, is the important step of body recovery, integration and consolidating memory.Sleep disorder
Refer to abnormal, sleep matter the exception of amount of sleep, or some clinical symptoms occur in sleep, sleep apnea (SAS) is exactly
A kind of common clinical symptoms.The symptom can not only have a strong impact on the sleep quality of patient, also result in generation hypertension, heart
The high-risk disease such as disease, cerebrovascular.And substantial amounts of sleep monitor equipment all has that with high costs, degree of accuracy is low, lets out on current market
The problems such as dew privacy and anti-interference difference, therefore, how to overcome problem above to realize that effective sleep monitor becomes in recent years
More popular research direction.Mainly there are following several sleep monitor means at present：
Wear auxiliary equipment.User needs to wear special auxiliary equipment, such as wrist strap, pectoral girdle, headband or probe etc., use
To gather the body data of user, so as to directly or indirectly be inferred to the sleep quality of the user.But these equipment one
As price it is high, need to force user to wear, and the monitoring information of coarseness can only be provided mostly, it is impossible to accurate supposition
The sleep quality of user.
Using various kinds of sensors.This kind of method relies on the array of pressure sensors being embedded in blanket, belt mostly, or
Vision, sound transducer etc..Monitoring method based on pressure transducer can provide fine-grained sleep quality information, but cost
It is too high.The monitoring method of other view-based access control models there is a problem of revealing privacy, and be affected larger by surrounding illumination intensity；
Monitoring method based on sound is for example using the iSleep of Doppler effect, environment larger by the effect of noise of surrounding
Robustness is poor.
Using the received signal strength (RSSI) of Wi-Fi.Radio signal can be not only used for transmission data, can also use
To perceive environment.Indoors under environment, the radio wave that signal transmitter is produced is via mulitpaths such as direct projection, reflection, scatterings
Propagate, multipath superposed signal is formed at signal receiver.Multipath superposed signal is propagated physical space by it to be affected, and can be taken
Information with reflection environmental characteristic.Liu et al. is in 2014 in document《Wi-Sleep:Contactless sleep
monitoring via Wi-Fi signals》In propose a kind of received signal strength of utilization Wi-Fi and carry out sleep monitor
Method, this process employs the periodic relationship between user's respiratory frequency and RSSI value reach monitoring user breathing shape
The purpose of state.Yet with the limitation of RSSI itself, such as RSSI can be because of little yardstick shade caused by signal multipath transmisstion
Decline and no longer with propagation distance monotone variation, meanwhile, under model experiment room environmental, a static receiver was at 1 minute
The RSSI for inside receiving is likely to occur the big ups and downs of 5dB so that the stability of this kind of method is not high.
In sum, at present in the urgent need to it is a kind of being capable of effective monitoring sleep quality and cost is not high under complex environment
Method.
The content of the invention
The invention aims to solve at present that monitoring sleep quality stability is not high under complex environment or cost
Too high problem, proposes a kind of method based on channel condition information monitoring sleep under complex environment.
Idea of the invention is that gathering the Wi-Fi channel condition informations in target environment by existing Wi-Fi equipment
(CSI) judge whether have mobile object to disturb around user, and according to CSI, if there is mobile object to delete disturbed information,
So as to obtain accurate respiration information, the environmental robustness of this sleep monitor method is greatly strengthen；In conjunction with principal component analysiss
Method and slip serial ports strategy, can effectively obtain the respiratory frequency of user, and by analyzing the respiratory frequency of user user is judged
Sleep quality.The inventive method is particularly suited for single sleep user, the complicated monitoring of environmental of multiple mobile objects.
The purpose of the present invention is achieved through the following technical solutions：
A kind of method under complex environment based on channel condition information monitoring sleep, comprises the following steps：
Step one, the CSI obtained in a period of time T range, adopt and go the peeling off in the original CSI of value filter removal that peel off
Value；
Step 2, for the CSI that step one is obtained, its phase information is processed using the method for linear transformation, obtain
Phase information after the calibration for obtaining becomes the effective form that can be used for mobile detection；
The CSI information that step 3, process step one and step 2 are obtained, according to time window length t and sliding window Δ t
Extract in T timeThe eigenvalue of the correlation matrix of CSI amplitudes and phase place in individual time window, then from CSI
Extract the CSI characteristic vectors in one time window of n composition in the eigenvalue of the correlation matrix of amplitude and phase place respectively,
CSI characteristic vectors in the time window are input to into the good grader of training in advance, classification results is obtained to judge whether
There is mobile object；If there is mobile object, the time window is marked；If without mobile object, not being marked；
Step 4, the CSI obtained for step one, using low pass filter and PCA height therein is removed
Frequency noise, acquisition can represent the information of user's breathing state, and using the breath signal breathing rate of user is calculated, and remove
Data in wherein markd time window.
Beneficial effect
The present invention only relies on common Wi-Fi equipment and collects Wi-Fi channel condition information CSI, it is possible to realize mobile object
Detection and breathing rate are calculated, so as to realize the monitoring of user's sleep quality.Therefore the present invention do not rely on various kinds of sensors and
Wearable device, low cost, strong interference immunity, to there is no leakage privacy concern, Consumer's Experience good, it is adaptable to single sleep user,
The complicated monitoring of environmental of multiple mobile objects.
Additionally, the present invention judges whether have mobile object to disturb around user according to CSI, if there is mobile object to delete
Disturbed information, greatly strengthen the environmental robustness of this sleep monitor method；In conjunction with PCA and slip serial ports
Strategy, can effectively obtain accurate respiration information, so that the inventive method has very high accuracy.
Description of the drawings
Fig. 1 is embodiment of the present invention sleep monitor method flow diagram.
Fig. 2 is that the embodiment of the present invention is used for monitoring of respiration accuracy rate of the environment without different user during mobile object.
Fig. 3 is the monitoring of respiration standard that the embodiment of the present invention is used for different user when environment has the object of different rate travels
True rate.
Fig. 4 is that the monitoring of respiration of same user when the embodiment of the present invention has multiple mobile objects for environment is accurate
Rate.
Specific embodiment
The preferred embodiment of the present invention is described further with reference to Figure of description and embodiment.
This method is a kind of method of monitoring user's sleep quality, can exclude in target environment and there is mobile object and bring
Interference, it is ensured that the user's breath signal for monitoring is efficiently and accurately, so as to judge the time period by breath signal
The height of the sleep quality of interior user.The handling process of this method is as shown in Figure 1.
For achieving the above object, the process that implements of the inventive method is comprised the following steps：
Step one, the CSI obtained in the T seconds, using the outlier for going to peel off in the original CSI of value filter removal.
Original CSI includes many outliers.Breathing and daily walking due to people etc. is a kind of low-frequency behavior, therefore
These outliers are more because caused by agreement itself and environment noise.These outliers can affect monitoring of respiration and shifting
Dynamic monitoring, therefore the first step processed original CSI is exactly outlier, the detailed process in the present embodiment is as follows：
The packet that first receiving terminal was received in the range of a period of time is common M, and a packet includes one group of CSI measurement
Value, includes m subcarrier in every group of CSI.Then the intermediate value μ and intermediate value for calculating original CSI sub-carriers amplitude is absolute
Deviations.Finally the value filter that peels off is gone using Hampel, remove it is any fall point outside [μ-γ σ, μ+γ σ], wherein γ is
One independent parameter, is set to 3 in the present embodiment.Certainly, as there is choosing, the present embodiment employs above-mentioned Hampel and removes outlier
Wave filter, one skilled in the art will appreciate that not limited to this, can be peeled off using other any methods or instrument to remove herein
Value.
CSI after step 2, the removal outlier that step one is obtained, using the method for linear transformation to its phase place letter
Breath carries out processing the phase information after being calibrated, and the information becomes the effective form that can be used for mobile detection.
Existing commercial wireless network card can not obtain effective phase information mostly, due to random noise and receiving terminal and
The clock of transmitting terminal is asynchronous, and original phase information becomes have no rule, it is difficult to directly utilize.Therefore need by linear transformation
Method obtain available phase information.Preferably, the present embodiment passes through procedure below by the subcarrier of all CSI in the T seconds
Original phase information be transformed into available form：
Use firstThe measured value of the phase place of i-th subcarrier is represented, wherein, φi
The true phase information of i-th subcarrier is represented, δ represents the time error for receiving extreme direction, and β is unknown phase deviation, and Z is
Measurement error.kiIt is the index value of i-th subcarrier of correspondence, wherein 1≤i≤m, in the present embodiment, has m=30 strips and carry
Ripple, corresponding index value is respectively：- 28, -26, -24, -22, -20, -18, -16, -14, -12, -10, -8, -6, -4, -2, -1,
1,3,5,7,9,11,13,15,17,19,21,23,25,27,28.N is the points that fast Fourier transform is adopted, and is set to 64.
Then makeFor two intermediate variables, because
ForCan regard as and be approximately 0, soThen calculate The linear transformation of true phase is obtained, the phase information after the calibration compares
It is more stable in measurement PHASE DISTRIBUTION.
Step 3, process the CSI information obtained Jing after step 2 phase alignment, according to default time window length t and
Sliding window Δ t extracts the eigenvalue of the correlation matrix of the CSI amplitudes in each time window and phase place, takes respectively
Maximum n eigenvalue cluster therein constitutes altogether the characteristic vector of the vector as CSI information in the window of a 2n*1, so
Afterwards this feature vector is input to into the good grader of training in advance, obtains classification results to determine whether mobile object；
If there is mobile object, the time period residing for window CSI information is marked；If without mobile object, not to the window
Time period residing for CSI information is marked.
For the CSI Jing after step one process, the cunning that a length of window is t seconds and each slip Δ t seconds is made first
Dynamic window, then in the T seconds one haveIndividual sliding window, it is assumed that k group CSI measured values are included in the window of this t second CSI, for
K group CSI measured values in each window, with the vectorial A of m × 1jThe amplitude of m subcarrier in represent jth group CSI measured value,
In order to eliminate the impact of signal absolute energy, need to all AjIt is standardized, that is, uses AjIn each element divided by | Aj| after
As a result newtonium is replaced, the result after standardization is designated as
Then useTo represent the correlation coefficient of two amplitude vectors, wherein 1≤i, j≤k,
By the correlation matrix for being calculated amplitude data
Then the feature of the matrix is calculated
Value, 3 eigenvalues for selecting maximum are designated as respectively a1, a2, a3。
Next, using process same above, for the CSI Jing after step 2 process, for k groups CSI in window
Measured value, with the vectorial φ of m × 1iThe phase place of m subcarrier in represent i-th group of CSI measured value, then to all φiCarry out
Standardization, the result after standardization is designated as
According to calculating ρAMethod calculate phase data correlation matrix ρφ, then calculate the feature of the matrix
Value, 3 eigenvalues for selecting maximum are designated as respectively c1, c2, c3。
The a that said process is obtained1, a2, a3, c1, c2, c3Vectorial Fea=[a that composition is one 6 × 11,a2,a3,c1,c2,
c3], vectorial Fea is exactly the characteristic vector of the CSI in this window.According to the CSI in one window of above-mentioned acquisition feature to
The method of amount, calculated in the T secondsIndividual Fea.
Multigroup CSI characteristic vectors with " whether having mobile object " label are produced according to the method described above as training sample
This, using algorithm of support vector machine a grader is set up.Finally by the characteristic vector of the CSI in a window for needing identification
The grader is input into, the label of " having mobile object " or " without mobile object " is obtained.If the label of a certain Fea is " have motive objects
Body ", flag F ea corresponding sliding window time；If the label of a certain Fea is " without mobile object ", do not make marks.
The method of above-mentioned acquisition correlation matrix and characteristic vector Fea is a preferred embodiment party of the present embodiment
Formula, one skilled in the art will appreciate that correlation matrix and Fea can be obtained using any other effective manner herein, only
Can be identified for that out the feature of the CSI.
Step 4, the CSI obtained for step one, using low pass filter and PCA height therein is removed
Frequency noise, acquisition can represent the information of user's breathing state, and using the breath signal breathing rate of user is calculated, and remove
There are the data in labelling time window.
Preferably, the present embodiment filters off the high frequency of each carrier wave in CSI first by Butterworth lowpass filters
Noise, then continues the breathing speed that denoising obtains user in the T seconds to the CSI after low-pass filtering with following PCAs
Rate：
The sliding window that length of window is t seconds and each slip Δ t seconds is made first, then one had in the T secondsIt is individual
Sliding window.Assume comprising k group CSI measured values in this t seconds window, with the vectorial HP of k × 1iAfter representing low-pass filtering
The time serieses of i-th subcarrier, the then time serieses of the m subcarrier matrix H P=[HP of k × m1,HP2,HP3,…,
HPm] represent；Then each element for the every string allowed in HP deducts the meansigma methodss of the row, after obtaining standardization
CalculateK × k covariance matrix S, and obtain the eigenvalue of S, then select the spy of corresponding k × 1 of wherein l eigenvalue
Levy vectorial e1、e2、……ei……el；Finally according to formulaCalculate k × 1
Breath signal breathSignal, wherein αiIt is weight coefficient, 1≤i≤l.Preferably, l is arranged in the present embodiment for maximum
2 eigenvalues, αiAll it is 0.5,1≤i≤2.
For the breath signal breathSignal obtained in current window, extract first wherein with regard to the letter of amplitude
Breath, plots oscillogram, and the interval between two crests is exactly user's breathing time once.Then by calculatingObtain the breathing cycle E, wherein p of useriIt is i-th pair adjacent wave peak value time interval, u
It is adjacent peaks logarithm, then breathing rate of the user within this t second is obtained by calculating R=60/E, the T seconds may finally be obtained
It is interiorIndividual breathing rate.Finally check whether each window corresponding time is labeled, it is such as labeled, then show this time
Internal respiration detection is interfered, the breathing rate that calculates is inaccurate, removes this time corresponding breathing rate, such as not
It is labeled, then retain.It is whether normally high to judge sleep quality of the user within this T second finally according to the breathing rate for retaining
It is low.
Embodiment
In order to test the performance of this method, from the room of 1 30 square meter, 1 TP-link router as transmitting terminal, 1
Notebook computer of the platform equipped with the network interface cards of Intel 5300 is used as receiving terminal, and the participant that 5 heights, age, body weight are different
Used as user, the packet sending speed of transmitting terminal is always every 20ms and sends a packet.
Monitoring of respiration accuracy rate of this method under noiseless situation is tested first.The length for arranging sliding window is 20 seconds
The packet number of interior acquisition, each sliding distance is 5 seconds.1 user lies flat on the bed, and another 1 user is quiet to stand and away from flat
Lie user and record lies low user's respiratory frequency of 5 minutes.Fig. 2 shows that different user lies low the experimental result of breathing, breathing
The accuracy rate of detection refer to the number of the breathing rate being consistent with true breathing rate calculated with this method in 5 minutes and
The total ratio of the breathing rate for calculating, the breathing rate that each window calculation goes out can regard the window intermediate time as
Breathing rate.As seen from Figure 2 the accuracy rate of monitoring of respiration has absolutely proved this method about between 84%～96%
There is higher accuracy.
Then capacity of resisting disturbance of this method when there is single mobile object is tested.The length for arranging sliding window is 1 second,
Every time sliding distance is 1 second, and 50 packets are included in a window.1 user lies flat on the bed, and another 1 user exists respectively
Quick, middling speed in room, move slowly at 5 minutes, 1 user is quiet vertical and away from the user and record lies low user at this 5 points of lying low
Respiratory frequency in clock.Fig. 3 shows the experimental result during object that there are different rate travels, as seen from the figure breathing prison
The accuracy rate of survey illustrates that this method is to the speed of mobile object and insensitive, either slowly about between 87%～96%
Speed walking or quick walking, can be judged with higher accuracy rate.
Capacity of resisting disturbance of the last test this method when there is multiple mobile objects.The length for arranging sliding window is 1 second,
Every time sliding distance is 1 second, and 50 packets are included in a window.1 user lies flat on the bed, have respectively 1,2,3
Name user is walked 5 minutes in room with phase same rate, and 1 user is quiet vertical and away from the user and record lies low user at this of lying low
Respiratory frequency in 5 minutes.Fig. 4 shows the experimental result that the user of varying number moves indoors, exhales as seen from the figure
The accuracy rate of monitoring is inhaled about between 89%～95%, illustrate this method no matter there is single mobile object still in environment
Multiple mobile objects can work well.
Chest regular fluctuating when being breathed due to user, causes the regular changes of CSI, is in particular in the width of CSI
Degree is presented sine wave state, and by analyzing the amplitude information of CSI the sleeping respiration information of user can be just obtained, and to the change of CSI
It is exactly other mobile objects in room to change topmost interference factor, therefore this method with the addition of mobile detecting step to exclude
The interference of mobile object, while including a large amount of subcarriers in CSI, all subcarriers are entered with process can effectively reduce due to a few
The error for producing is forbidden in individual subcarrier measurement, it is ensured that the higher stability of this method.In sum, the present invention is in single sleep
In user, the complex environment of multiple mobile objects, the breathing state and sleep quality of user can be accurately and effectively monitored.
In order to illustrate present disclosure and implementation, this specification gives specific embodiment.Draw in embodiment
The purpose for entering details is not to limit the scope of claims, and is to aid in understanding the method for the invention.The technology of this area
Personnel should be understood that：In without departing from the present invention and its spirit and scope of the appended claims, to each of most preferred embodiment step
It is all possible to plant modification, change or replacement.Therefore, the present invention should not be limited to most preferred embodiment and interior disclosed in accompanying drawing
Hold.
Claims (10)
1. the method for channel condition information monitoring sleep is based under a kind of complex environment, it is characterised in that comprised the steps：
Step one, obtains the Wi-Fi channel condition information CSI in a period of time T range, adopts and goes the value filter removal original that peels off
Outlier in beginning CSI；
Step 2, for the CSI that obtained Jing after step one process, using the method for linear transformation to its phase information at
Reason, the phase information after the calibration of acquisition becomes the effective form that later step can be used for mobile detection；
Step 3, processes the CSI information obtained Jing after step one and step 2 are processed, according to time window length t and sliding window
Mouth Δ t is extracted in T timeThe eigenvalue of the correlation matrix of CSI amplitudes and phase place in individual time window, then
Extract the CSI features in one time window of n composition respectively from the eigenvalue of CSI amplitudes and the correlation matrix of phase place
Vector, by the CSI characteristic vectors in the time window the good grader of training in advance is input to, and obtains classification results to judge
Whether mobile object is had：If there is mobile object, the time window is marked；If without mobile object, not carrying out
Labelling；
Step 4, for the CSI obtained Jing after step one process, is removed wherein using low pass filter and PCA
High-frequency noise, acquisition can represent the information of user's breathing state, using the breath signal breathing rate of user is calculated, and
Remove the data in wherein markd time window.
2. the method for channel condition information monitoring sleep, its feature are based under a kind of complex environment according to claim 1
It is：The value filter that goes to peel off removes the value filter that peels off for Hampel.
3. the method for channel condition information monitoring sleep, its feature are based under a kind of complex environment according to claim 1
It is：Method described in step 2 using linear transformation is as follows to detailed process that its phase information is processed：
First, useThe measured value of the phase place of i-th subcarrier of CSI is represented, wherein, φi
Represent the true phase information of i-th subcarrier, δ represents the time error for receiving extreme direction, β is unknown phase deviation, Z tables
Show measurement error, N represents the points that fast Fourier transform is adopted；
Then makeFor two intermediate variables, becauseCan regard as and be approximately 0, so
Finally calculateObtain true phase
Linear transformation, it is more stable that the phase information after the calibration is compared to measurement PHASE DISTRIBUTION.
4. the method for channel condition information monitoring sleep, its feature are based under a kind of complex environment according to claim 1
It is：The correlation matrix of CSI amplitudes and phase place in the one time window of step 3 passes through following Procedure Acquisitions：
First, remember that the CSI in a time window includes k group measured values；
Secondly, to this k group measured value, with the vectorial A of m × 1jThe amplitude or phase of m subcarrier in represent jth group CSI measured value
Position, wherein 1≤j≤k, in order to eliminate the impact of signal absolute energy, to all AjIt is standardized and obtains
Finally, useTo represent the correlation coefficient of two amplitude vectors or two phase vectors, its
In 1≤i, j≤k, by the correlation matrix for being calculated amplitude data or phase data
5. the method for channel condition information monitoring sleep, its feature are based under a kind of complex environment according to claim 1
It is：N described in step 3 is maximum 3.
6. the method for channel condition information monitoring sleep, its feature are based under a kind of complex environment according to claim 1
It is：Low pass filter described in step 4 is Butterworth lowpass filters.
7. the method for channel condition information monitoring sleep, its feature are based under a kind of complex environment according to claim 1
It is：Being obtained using the high-frequency noise in PCA removal CSI described in step 4 can represent user's breathing state
The detailed process of information is as follows：
First, remember that the CSI in a time window includes k group measured values；
Then, with the vectorial HP of k × 1iThe time serieses of i-th subcarrier of CSI are represented, then the time sequence of m subcarrier
The row matrix H P=[HP of k × m1,HP2,HP3,…,HPm] represent；Each element for the every string allowed in HP deducts this
The meansigma methodss of row, after obtaining standardizationCalculateK × k covariance matrix S, and obtain the eigenvalue of S, select it
Characteristic vector e of corresponding k × 1 of middle n eigenvalue1、e2、……ei……en；Wherein 1≤i≤n；
Finally, according to formulaThe breath signal breathSignal of k × 1 is calculated,
Wherein αiIt is weight coefficient, 1≤i≤n.
8. the method for channel condition information monitoring sleep, its feature are based under a kind of complex environment according to claim 7
It is：The n eigenvalue is 2 maximum eigenvalues.
9. the method for channel condition information monitoring sleep, its feature are based under a kind of complex environment according to claim 7
It is：It is described
10. according to the method under a kind of arbitrary described complex environment of claim 1-9 based on channel condition information monitoring sleep,
It is characterized in that：The detailed process for calculating the breathing rate of user described in step 4 using breath signal is as follows：During for one
Between the breath signal breathSignal that obtains in window, extract first wherein with regard to the information of amplitude, plot oscillogram, two
Interval between individual crest is exactly user's breathing time once；Then by calculatingObtain
Obtain the breathing cycle E, wherein p of useriIt is i-th pair adjacent wave peak value time interval, u is adjacent peaks logarithm；Finally by meter
Calculate R=60/E to obtain breathing rate of the user in this time window.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611145695.8A CN106618497A (en) | 2016-12-13 | 2016-12-13 | Method for monitoring sleep in complicated environment based on channel state information |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611145695.8A CN106618497A (en) | 2016-12-13 | 2016-12-13 | Method for monitoring sleep in complicated environment based on channel state information |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106618497A true CN106618497A (en) | 2017-05-10 |
Family
ID=58824431
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611145695.8A Pending CN106618497A (en) | 2016-12-13 | 2016-12-13 | Method for monitoring sleep in complicated environment based on channel state information |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106618497A (en) |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107333105A (en) * | 2017-07-14 | 2017-11-07 | 中国矿业大学 | A kind of downhole intelligent video monitoring system and monitoring method based on WIFI |
| CN107518873A (en) * | 2016-06-22 | 2017-12-29 | 北京小米移动软件有限公司 | The method and device of monitoring sleep situation, electronic equipment |
| CN107633227A (en) * | 2017-09-15 | 2018-01-26 | 华中科技大学 | A kind of fine granularity gesture identification method and system based on CSI |
| CN107862260A (en) * | 2017-10-26 | 2018-03-30 | 西北大学 | A kind of target identification method |
| CN107968689A (en) * | 2017-12-06 | 2018-04-27 | 北京邮电大学 | Perception recognition methods and device based on wireless communication signals |
| WO2019080728A1 (en) * | 2017-10-23 | 2019-05-02 | 叶伟 | Method for detecting heart rate abnormality on the basis of wi-fi signal |
| CN109745026A (en) * | 2017-11-07 | 2019-05-14 | 深圳欧德蒙科技有限公司 | A kind of method for measuring heart rate and system |
| CN110123328A (en) * | 2019-06-26 | 2019-08-16 | 南京苗米科技有限公司 | A kind of respiratory rate detection method based on wireless identification |
| CN111225354A (en) * | 2020-02-14 | 2020-06-02 | 重庆邮电大学 | CSI human body falling identification method under WiFi interference environment |
| CN111568425A (en) * | 2020-06-08 | 2020-08-25 | 北京大学 | Non-contact type multi-person respiration detection method |
| CN111770705A (en) * | 2017-12-28 | 2020-10-13 | 数眠公司 | Bed with presence detection feature |
| CN114916912A (en) * | 2022-05-09 | 2022-08-19 | 大连理工大学 | Non-contact sleep apnea detection method and device |
| CN115040092A (en) * | 2022-06-13 | 2022-09-13 | 天津大学 | Heart rate monitoring method and respiratory event detection method based on channel state information |
| WO2023241469A1 (en) * | 2022-06-13 | 2023-12-21 | 华为技术有限公司 | Sleep monitoring method and device |
| WO2024002029A1 (en) * | 2022-06-30 | 2024-01-04 | 华为技术有限公司 | Respiration test method, and electronic device, storage medium and program product |
-
2016
- 2016-12-13 CN CN201611145695.8A patent/CN106618497A/en active Pending
Non-Patent Citations (1)
| Title |
|---|
| LI FAN: "Mo-sleep: Unobtrusive sleep and movement monitoring via Wi-Fi signal", 《2016 IEEE 35TH INTERNATIONAL PERFORMANCE COMPUTING AND COMMUNICATIONS CONFERENCE (IPCCC)》 * |
Cited By (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107518873A (en) * | 2016-06-22 | 2017-12-29 | 北京小米移动软件有限公司 | The method and device of monitoring sleep situation, electronic equipment |
| CN107333105A (en) * | 2017-07-14 | 2017-11-07 | 中国矿业大学 | A kind of downhole intelligent video monitoring system and monitoring method based on WIFI |
| CN107633227A (en) * | 2017-09-15 | 2018-01-26 | 华中科技大学 | A kind of fine granularity gesture identification method and system based on CSI |
| CN107633227B (en) * | 2017-09-15 | 2020-04-28 | 华中科技大学 | CSI-based fine-grained gesture recognition method and system |
| WO2019080728A1 (en) * | 2017-10-23 | 2019-05-02 | 叶伟 | Method for detecting heart rate abnormality on the basis of wi-fi signal |
| CN107862260A (en) * | 2017-10-26 | 2018-03-30 | 西北大学 | A kind of target identification method |
| CN107862260B (en) * | 2017-10-26 | 2021-06-04 | 西北大学 | Target identification method |
| CN109745026A (en) * | 2017-11-07 | 2019-05-14 | 深圳欧德蒙科技有限公司 | A kind of method for measuring heart rate and system |
| CN107968689A (en) * | 2017-12-06 | 2018-04-27 | 北京邮电大学 | Perception recognition methods and device based on wireless communication signals |
| CN111770705A (en) * | 2017-12-28 | 2020-10-13 | 数眠公司 | Bed with presence detection feature |
| CN111770705B (en) * | 2017-12-28 | 2023-06-02 | 数眠公司 | Bed with presence detection feature |
| CN110123328B (en) * | 2019-06-26 | 2022-04-15 | 南京苗米科技有限公司 | Breathing frequency detection method based on wireless identification |
| CN110123328A (en) * | 2019-06-26 | 2019-08-16 | 南京苗米科技有限公司 | A kind of respiratory rate detection method based on wireless identification |
| CN111225354B (en) * | 2020-02-14 | 2022-02-22 | 重庆邮电大学 | CSI human body falling identification method under WiFi interference environment |
| CN111225354A (en) * | 2020-02-14 | 2020-06-02 | 重庆邮电大学 | CSI human body falling identification method under WiFi interference environment |
| CN111568425A (en) * | 2020-06-08 | 2020-08-25 | 北京大学 | Non-contact type multi-person respiration detection method |
| CN111568425B (en) * | 2020-06-08 | 2021-08-03 | 北京大学 | Non-contact type multi-person respiration detection method |
| CN114916912A (en) * | 2022-05-09 | 2022-08-19 | 大连理工大学 | Non-contact sleep apnea detection method and device |
| CN115040092A (en) * | 2022-06-13 | 2022-09-13 | 天津大学 | Heart rate monitoring method and respiratory event detection method based on channel state information |
| WO2023241469A1 (en) * | 2022-06-13 | 2023-12-21 | 华为技术有限公司 | Sleep monitoring method and device |
| WO2024002029A1 (en) * | 2022-06-30 | 2024-01-04 | 华为技术有限公司 | Respiration test method, and electronic device, storage medium and program product |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106618497A (en) | Method for monitoring sleep in complicated environment based on channel state information | |
| CN106725488B (en) | Wireless field intensity respiration detection method and device and respiration detector | |
| Liu et al. | Contactless respiration monitoring via off-the-shelf WiFi devices | |
| CN109998549A (en) | A kind of human body respiration detection method based on WiFi channel state information | |
| Xu et al. | WiStep: Device-free step counting with WiFi signals | |
| CN105451652B (en) | For determining the processing unit and processing method of the breath signal of object | |
| CN111568425B (en) | Non-contact type multi-person respiration detection method | |
| CN102917661A (en) | Multivariate residual-based health index for human health monitoring | |
| CN112386236A (en) | Vital sign monitoring method based on channel state phase information | |
| CN109431465B (en) | Old people sleep quality monitoring method based on radio frequency tag technology | |
| Li et al. | Mo-sleep: Unobtrusive sleep and movement monitoring via Wi-Fi signal | |
| CN103006195A (en) | Non-contact vital sign data monitoring system and non-contact vital sign data monitoring method on basis of image processing | |
| Zhang et al. | Wicare: Towards in-situ breath monitoring | |
| WO2019080728A1 (en) | Method for detecting heart rate abnormality on the basis of wi-fi signal | |
| CN107405106A (en) | Respiration rate detection means, respiration rate detection method and program recorded medium | |
| CN109330597A (en) | A kind of human body respiration method for tracing based on channel state information | |
| Liu et al. | Human biometric signals monitoring based on wifi channel state information using deep learning | |
| CN108992053A (en) | A method of real-time chainless detection heart rate and eartbeat interval | |
| CN113743374B (en) | Personnel identity recognition method based on channel state information respiratory perception | |
| Muaaz et al. | WiHAR: From Wi-Fi channel state information to unobtrusive human activity recognition | |
| CN206979487U (en) | A kind of wireless field density breathing detection instrument | |
| CN110338798B (en) | Human body respiration quantity evaluation system and method based on RFID (radio frequency identification) tag | |
| CN110123328B (en) | Breathing frequency detection method based on wireless identification | |
| CN111839520A (en) | Human respiration monitoring method and device based on CSI signal power response autocorrelation | |
| Zhuang et al. | SleepSense: Non-invasive sleep event recognition using an electromagnetic probe |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20170510 |