CN108523901A - A kind of sleep quality monitoring method based on smart mobile phone - Google Patents
A kind of sleep quality monitoring method based on smart mobile phone Download PDFInfo
- Publication number
- CN108523901A CN108523901A CN201810455152.9A CN201810455152A CN108523901A CN 108523901 A CN108523901 A CN 108523901A CN 201810455152 A CN201810455152 A CN 201810455152A CN 108523901 A CN108523901 A CN 108523901A
- Authority
- CN
- China
- Prior art keywords
- sleep
- stage
- mobile phone
- sleeper
- quality
- 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
- A61B5/103—Detecting, measuring or recording devices for testing the shape, pattern, colour, size or movement of the body or parts thereof, for diagnostic purposes
- A61B5/11—Measuring movement of the entire body or parts thereof, e.g. head or hand tremor, mobility of a limb
- A61B5/1118—Determining activity level
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/08—Detecting, measuring or recording devices for evaluating the respiratory organs
- A61B5/0826—Detecting or evaluating apnoea events
-
- 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
- A61B5/4809—Sleep detection, i.e. determining whether a subject is asleep or not
-
- 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
-
- 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
- A61B5/4818—Sleep apnoea
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/68—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient
- A61B5/6801—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient specially adapted to be attached to or worn on the body surface
- A61B5/6802—Sensor mounted on worn items
- A61B5/681—Wristwatch-type devices
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/68—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient
- A61B5/6887—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient mounted on external non-worn devices, e.g. non-medical devices
- A61B5/6898—Portable consumer electronic devices, e.g. music players, telephones, tablet computers
Abstract
The sleep quality monitoring method based on smart mobile phone that the invention discloses a kind of, belongs to smart mobile phone mobile awareness technical field.First with the sleep relevant physiological characteristic signal in the sensor acquisition sleep procedure of mobile phone, then feature extraction is carried out to physiological characteristic signal, recycle fuzzy logic theory method by corresponding Feature Mapping to corresponding sleep state, age, the ambient lighting information for finally combining sleeper, provide sleep quality score.By comparing 20 groups of dormant datas, the judging nicety rate that can obtain REM reaches 75.3%, and the judging nicety rate of either shallow sleep stage reaches 81.2%, and the judging nicety rate of deep sleep reaches 78.8%.It can be seen that, the judging nicety rate highest of either shallow sleep, the judging nicety rate of rapid-eye-movement sleep (REM) is minimum, this is because similar with the cerebral neuron activity in awake stage in the REM stages, limb motion and respiratory rate can not effectively screen the two stages, be easy to cause erroneous judgement.
Description
Technical field
The invention belongs to smart mobile phone mobile awareness technical fields, utilize smart mobile phone to survey more particularly under indoor scene
The method for measuring user's sleep quality.
Background technology
With development in science and technology, people's living standard is gradually increased, and health problem is gradually concerned by people.Sleep is one
The very important physiological activity of item, the time in all one's life about 1/3 of people are spent in sleep, and the quality of sleep quality is straight
Connect the health for affecting people.Sleep insufficiency can lead to many adverse consequences, such as headache, anxiety, immunity degradation etc..
Therefore, the highly desirable sleep quality to oneself of people understands there are one intuitive.Medically use polysomnogram acquisition sleep
Eeg signal, the electrocardiosignal of person also has the physiological signals such as eye movement signal to judge sleeper's sleep quality, it is considered to be most
Accurate measurement method but due to measured needs to wear various monitoring devices the whole night, and needs professional to grasp
Make, so being not suitable for ordinary populace routine use.
Current sleep monitor equipment, focuses primarily upon Intelligent bracelet and sleeping bed cushion class product, wherein bracelet product with
Based on Fitbit, Jawbone UP and millet bracelet, this kind of equipment mainly using the body movement signal of sleeper's sleep procedure come pair
Sleep state is judged, is not acquired analysis for other sleep coherent signal especially breath signals, therefore obtain
Dormant data can not completely embody the sleep quality of sleeper, for example snore and can not just be judged, and using process
In need to wear the whole night, sense of discomfort can be brought to user, a good user experience is not provided.Shut-eye bed mattress product passes through
The sensor being embedded in mattress acquires the body movement signal in user's sleep procedure, breath signal even heartbeat signal to sentence
Disconnected sleep state, measurement result is more accurate, but this kind of product is generally more expensive, and since build is larger, makes
It uses nor easily.
It is good for so user carrys out the sleep to oneself there is an urgent need to a easy to use, the comprehensive sleep monitor product of measurement
The effective management of Kang Jinhang mono-.
Invention content
For defect existing for existing sleep monitor method, the present invention proposes a kind of sleep quality based on smart mobile phone
Monitoring method.Different from previous methods, the present invention is when measuring sleeper's sleep quality, it is only necessary to will be equipped with sleep prison
The mobile phone of examining system is placed on user's bedside, so that it may to complete sleep monitor work.The present invention has considered in sleep procedure
Body movement signal, breath signal judge sleep state, sleep then in conjunction with age of user and ambient lighting information to user
Quality is given a mark, and bracelet equipment is different from, and the present invention can also measure snoring, and user is allowed to be found as early as possible with sleep
Suspend the Potential feasibility of syndrome.
The technical solution adopted by the present invention is a kind of sleep quality monitoring method based on smart mobile phone, first with mobile phone
Sleep relevant physiological characteristic signal in included sensor acquisition sleep procedure, then carries out feature to physiological characteristic signal and carries
It takes, recycles fuzzy logic theory method by corresponding Feature Mapping to corresponding sleep state, finally combine the year of sleeper
Age, ambient lighting information, provide sleep quality score, and overall implementation is as shown in Figure 1.
It is specifically divided into following steps：
The dynamic monitoring of 1.S1 bodies
In the different phase of sleep, sleeper can show different limb motion behaviors, in either shallow sleep stage
(Light Sleep) amplitude of being easy to happen changes limb motion greatly, for example turn over etc., in deep sleep stages (Deep
Sleep), sleeper is not easy to be waken, and heart rate slows down, and changes big limb motion to be not susceptible to amplitude, in rapid eye movement
The E.E.G of sleep stage (REM), sleeper is in fast transition state, often will appear casual leg extension, arm is waved
Phenomena such as.
Body movement information is obtained using the acceleration transducer built in smart mobile phone, it, will in order to obtain accurate body movement information
The sample frequency of acceleration transducer built in smart mobile phone is set as 100Hz, and the data of acceleration transducer acquisition are divided into X
Three axis, Y-axis and Z axis directions, use a respectivelyx(i)、ay(i)、az(i) X-axis is indicated, i-th on three directions of Y-axis and Z axis
Then a sample value calculates total acceleration a (i) with formula (1), recycle formula (2) to carry out acceleration change V (i) later
Calculating.
V (i)=a (i)-a (i-1) (2)
By way of threshold value is arranged, to determine whether there is the generation that body moves state.If the acceleration value V measured
(i) ＞ ε, being considered as current acceleration variation is moved caused by state change by body, if the opposite acceleration value V measured
(i) current value variation is then attributed to noise and caused by≤ε, and ε is the threshold value of setting；By being tested to 10 volunteers
ε, is set as 0.05 by analysis.
The type of action of sleeper is divided into macro motion and micro motion, is judged according to the length of action time of origin.
Because from experiment it can be found that the big small limb motion of limb motion and amplitude of variation of amplitude of variation is deposited on the duration
In apparent difference.The duration of big-movement is average, and all in 1s or more, the duration of little trick is both less than 1s, so if
It measures limb motion duration and is less than 1s, it is little trick to be considered as the limb motion.If being more than conversely, measuring limb motion duration
The limb motion is just classified as big-movement by 1s.
S2 monitoring of respiration
Monitoring of respiration is judged the respiratory rate of sleeper, because in different sleep stages, the breathing of sleeper
There are significant differences for frequency, and at rapid-eye-movement sleep stage (REM), E.E.G variation is rapid, and it is quick that sleeper often will appear eye
Jumping phenomena is simultaneously had a dream in this stage, so the respiratory rate in this stage sleep person is usually unstable.In either shallow sleep rank
Section (Light Sleep), eye jumping phenomena stop, and sleeper's respiratory rate tends to be steady.In deep sleep stages (Deep
Sleep), the respiratory rate of sleeper further slows down.So the size of respiratory rate with the sleep stage residing for sleeper not
It is same and different.
Microphone acquires voice data according to the sample frequency of 8KHz, and voice data is one group according to every 400 sample points
N group data are divided into, N indicates the group number of sample point；Because the sample frequency being arranged is 8kHz, 400 samples
The time interval of point is 0.05s, then sets sample data in every group to the average value of this group of data, the purpose done so
It is the complexity that will be calculated after reducing.According to the periodic feature of breath signal, the data after a cycle have similar
Property, so for first of sample point e (l), e (l)=e (l+T) is obtained, T indicates the sample point number in a cycle.Normally
Respiration rate when adult's peace and quiet is within 16 to 20 times, so setting time minimum period Tmin=16000, maximum week
Time phase Tmax=40000 do a limitation to cycle time, and calculation formula (3) is as follows：
Allow T in TminAnd TmaxIncrease in range, finds corresponding T values when functional value minimum, this makes it possible to calculate
The duration t=T/8000 of a cycle is obtained, corresponding respiratory rate is exactly 60/t.Step-length when formula (3) calculates
400 are set as, because the data in each group are set as the average value of this group of data, calculative number subtracts significantly
Few, since this system operates on smart mobile phone, system power dissipation can be reduced by mitigating calculation amount.
The S3 sounds of snoring monitor
Snoring has certain common features with Breathiness, as shown in Fig. 2, being Breathiness and the sound of snoring respectively.From figure
As can be seen that Breathiness has similitude with snoring sound on waveform, because they all have periodically, difference exists
It is bigger than the mean amplitude of tide of Breathiness in the sound of snoring, this is because the reason that the sound of snoring is usually bigger than Breathiness.Again from spectrogram
It has been observed that the two frequency distribution has significant difference, the frequency distribution of the sound of snoring concentrates on 0 and arrives the low-frequency range of 1000Hz, and exhales
Sound absorption sound be then all distributed in low frequency and high band it is more, therefore can according to the ratio of low-frequency range and high band come judge acquisition
Sample sound is Breathiness or the sound of snoring.
Judge the presence of the sound of snoring using the prevailing frequency range of current sound sample segment, is indicated containing n with f
The voiced frame to be measured of a sample point, siF i-th of data value after Fourier changes is indicated, then allowing low-frequency range and high band
Ratio r lh with formula (4) indicate.
When sleeper is snored by eupnea state shift, sound frequency range will be tilted gradually to low frequency, be caused
Rlh values gradually increase.By analyzing the Snore data more than 100 parts of different sleepers, about settled low frequency and high frequency ratio are gradual
Become larger and rlh (f)>Just illustrate that active user is in snoring state when 1.5, it is on the contrary then be in eupnea state.It is logical
The duration of overwriting user snoring provides snoring duration data when providing sleep quality report, for reference.
S4 illumination monitors
Environment illumination intensity can also have an impact the sleep quality of sleeper, in general, environment illumination intensity compared with
Under strong environment, light can be uncomfortable by eyes through eyelid, poor so as to cause the sleep quality of sleeper.Conversely, in ring
Under the weaker environment of border intensity of illumination, the sleep quality of sleeper is usually preferable, so, environment illumination intensity is also with sleeper's
Sleep quality is closely bound up.
Carrying out there are problems that two when light sensor carries out intensity of illumination monitoring, first, light is generally installed
The purpose of sensor is when extraneous environment illumination intensity changes, and dynamic adjusts screen intensity, so light sensor is all installed
In mobile phone front.Mobile phone is close to the users during sleep monitor, and unconscious body kinematics of the sleeper in sleep procedure have
Handset may be caused to overturn or sheltered from by foreign object, if mobile phone front is blocked by foreign object, will result in light
Sensor reads environment illumination intensity data inaccuracy, and interference is generated to result.Second, intensity of illumination, which measures, to be needed with user
The whole process of sleep, if light sensor long-play is bound to cause the aggravation of electric quantity consumption, so to be arranged rationally
Operation method reach energy-efficient purpose.
For first problem above-mentioned, since mobile phone front light sensor is closely located to range sensor,
So can perceive whether mobile phone is blocked with range sensor, detected with the last time if perceiving mobile phone and being blocked
Light intensity analyzed as current intensity of illumination, when monitoring that mobile phone is not blocked, then detect light again
It is judged as low light environment when intensity of illumination is less than 10Lux according to intensity；When intensity of illumination, which is more than 10Lux, is less than 100Lux,
It is judged as medium light environment；When intensity of illumination is more than 100Lux, it is judged as strong light environment；Its flow can be indicated with Fig. 3.
For Second Problem, it is contemplated that intensity of illumination will not change especially frequent, so system does not need to real-time monitoring of environmental
Intensity of illumination.In the method, detection time interval is set to 10 minutes, can thus plays the mesh for saving mobile phone electricity
's.
S5 sleep states judge
The present invention carries out sleep state for the uncertain feature in sleep state prediction, with fuzzy logic operation method
Judge.Detailed process is shown in Fig. 4.
Detailed process is divided into following steps：
S5.1 collection analysis data.One section of continuous sleep physiological data is acquired, then using 30s as one-shot measurement interval,
The monitoring data for taking each 600s before and after this section of continuous sleep period, are slept as this section of continuous sleep period
The detection data of dormancy condition adjudgement.
S5.2 sleep stage fuzzy logic systems generate.This system is to be moved according to body and breathe this to the judgement of sleep stage
Two factors, so extraction does riding Quality Analysis in the average value of this section of continuous sleep period internal respiration frequency, then
Form is moved to body to compare, and is calculated the average time of macro motion and micro motion generation, is thus obtained fuzzy logic system
Input variable：
X1：The average value and stability of respiratory rate are divided into normal, relatively slow and slow.
X2：Body moves big-movement frequency, is divided into normal, less and few.
The output variable of system is exactly the three phases slept, i.e. F1：Rapid-eye-movement sleep stage (REM), F2：Either shallow is slept
Dormancy stage (Light Sleep), F3：Deep sleep stages (Deep Sleep).
Thus generating a tool, there are two input variable X1And X2With three output F1、F2、F3Related sleep stage
Fuzzy logic system.
S5.3 membership functions calculate.According to the input variable and output variable set, each input of Rational choice becomes
Then the membership function of amount carries out integration operation.Triangleshape grade of membership function is selected to carry out analytical Calculation, analytic expression is formula
(5), membership function image such as Fig. 5.
S5.4 generates fuzzy logic ordination.The characteristics of being changed using different physiological signals in sleep procedure forms fuzzy patrol
Rule is collected, specifically describing rule is：
S5.4.1. in sleep procedure respiratory rate change mechanism.
In REM or awake stage, respiratory rate is very fast and unstable, deep sleep stages breathing most slowly and
Steadily, it is then intervened therebetween in either shallow sleep stage, respiratory rate is relatively slow and relatively steady.
So the changing rule in each sleep stage respiratory rate is:It is awake>REM>Either shallow is slept>Deep sleep.
S5.4.2. in sleep procedure limb motion change mechanism.
The dynamic number of body is relatively more when awake and concentrates on big-movement, and the REM stages are easier that the big-movements such as turn occur,
Either shallow sleep stage big-movement is reduced, and is concentrated on this little trick of body twitch, has been arrived deep sleep stages, limb action is basic
It disappears, even if it is also slight little trick to have.
The variation rule that can obtain moving frequency according to body in sleep and size is ranked up according to above-mentioned judgement
Rule is：It is awake>REM>Either shallow is slept>Deep sleep.
After obtaining the above rule, operation is carried out by the membership function to different input variables, is then done final
As a result it exports.If it is few that body moves number, and is little trick, and respiratory rate is slow, then this period is for deep sleep
Degree of membership just be 1.
S5.5 anti fuzzy methods.Anti fuzzy method is carried out to the result of calculation that previous action obtains, can just obtain wanting result.
In the case where different input variables have different membership functions, the result of calculation with maximum membership degree is chosen as final
As a result.
Here it is carry out the overall of sleep judgement based on fuzzy logic theory to realize algorithm.
S6 sleep qualities judge
Sleeper is in sleep procedure the whole night, and the quality of sleep quality is not only related with total sleep time, also by each
The influence of sleep stage distribution proportion, the time scale of sleep quality judgment method of the invention dependent on each stage of sleeping, and
And userspersonal information and ambient lighting information are combined, the sleep quality report of individual subscriber is provided after comprehensive analysis.
With age, the sleeping time of people can continue to reduce, the standard sleep that each age bracket has oneself special
Time.For example, the daily sleep in 8 hours of young man in 20 to 30 years old is sufficient, 31 to 60 years old adults then need 6.49
Sleeping time hour, more time that women needs is 7.5 hours, the reason is that climacteric continuous sleep rhythm changes.
For 60 years old or more the elderly, became shorter and shorter sleeping time in the evening, and 7 hours even 5.5 hours were sufficient.So
It carries out sleep quality judgement and first determines whether sleep duration reaches respective standard.In this system, sleep quality point is represented with SQ
Number, full marks are 100 points, if reaching corresponding sleep duration standard, do not deduct points, certain point is otherwise subtracted on the basis of total score
Number detains 5 points per poor half an hour.
Intensity of illumination can also have an impact sleep quality, after the completion of each sleep procedure, when system monitoring to user
It when intensity of illumination is higher in sleep procedure, can also be calculated accordingly, preliminary intermediate light of arranging is according to 2 points of button, strong light button 5
Point.After finally summarizing deduction of points item, it is S to enable the gross score deducted.
Determine that another key factor of sleep quality is the Annual distribution of different sleep stages, the sleep of a high quality
Process should be that each sleep stage ratio is reasonable, and is not easy to be disturbed in sleep.Medical research finds that human body is in depth
Degree sleep stage can carry out the recovery of body function, and then contribute to the solidification of memory and space to remember in the rapid-eye-movement sleep phase
The formation recalled, therefore the reasonability and validity for wanting comprehensive consideration difference sleep stage just to can guarantee that sleep quality judges.It is examining
After considering each factor, sleep quality calculation formula finally is set as (6).
Sleep quality marking effect, exactly allow user find in time oneself sleep there are the problem of, from the sleep of oneself
In mass change trend, finds the factor for influencing oneself sleep, problem is investigated in time, so as to improve sleep quality.
Description of the drawings
Fig. 1 sleep qualities monitor flow chart.
Fig. 2 is breathed to be compared with the sound of snoring.
Intensity flow chart is looked after in Fig. 3 detections.
Fig. 4 sleep stage algorithm flow charts.
Fig. 5 Triangleshape grade of membership function images.
Fig. 6 sleep total duration comparisons.
Each stage monitoring accuracy of Fig. 7 sleeps.
Specific implementation mode
In the present invention, in order to ensure the accuracy of test result, have collected 20 volunteers altogether totally 60 days sleep numbers
According to three days dormant datas of every volunteer.To ensure the randomness of sampling, the age distribution of volunteer was from 20 years old to 60
Year, and it is divided into 10 males and 10 women.When every volunteer measures at the same wear Jawbone UP bracelets and
The smart mobile phone of sleep monitor system is installed, volunteer opens simultaneously the sleep monitor on bracelet and smart mobile phone in sleep
Function, bracelet are worn in wrist, and smart mobile phone is positioned over beside pillow, and system carries out dormant measurement, passes through automatically
Both is compared to compare the accuracy of this system measurement.
It is dormant to test the comparison for being divided into two levels, it is to measure comparison the different sleep stage times first, secondly
It is total sleep duration time of measuring comparison.The dormant data for randomly selecting one day to every volunteer from measurement data concentration, obtains
To the sleep duration comparison diagram being illustrated in fig. 6 shown below.
From experimental result as can be seen that the sleep total duration error mean that Jawbone UP bracelets and smart mobile phone measure exists
Within 50 minutes, the 12.3% of sleep mean value duration is accounted for, i.e., accurately reaches 87.7% with bracelet comparison average monitored.Experiment knot
Fruit shows that this system has very high accuracy in sleep total duration monitoring compared with bracelet product.
Sleep monitor is not only in that the record of sleep total duration, and what user more valued is the time point of different sleeping periods
Cloth, therefore by carrying out analysis comparison to experimental data set, the test result of Fig. 7 is obtained, the sleep state point measured with bracelet
Cloth is used as the standard of referring to.
By comparing 20 groups of dormant datas, the judging nicety rate that can obtain REM reaches 75.3%, either shallow sleep stage
Judging nicety rate reaches 81.2%, and the judging nicety rate of deep sleep reaches 78.8%.As can be seen that the judgement of either shallow sleep is accurate
True rate highest, the judging nicety rate of rapid-eye-movement sleep (REM) is minimum, this is because in the brain in REM stages and awake stage
Neuron activity is similar, and limb motion and respiratory rate can not effectively screen the two stages, be easy to cause erroneous judgement.
Claims (3)
1. a kind of sleep quality monitoring method based on smart mobile phone, it is characterised in that：First with the sensor of mobile phone
The sleep relevant physiological characteristic signal in sleep procedure is acquired, feature extraction then is carried out to physiological characteristic signal, recycles mould
Corresponding Feature Mapping to corresponding sleep state, is finally combined age, the ambient lighting of sleeper by fuzzy logic theoretical method
Information provides sleep quality score；
It is specifically divided into following steps：
The dynamic monitoring of S1 bodies
Obtaining body movement information using the acceleration transducer built in smart mobile phone will be intelligent in order to obtain accurate body movement information
The sample frequency of acceleration transducer built in mobile phone is set as 100Hz, and the data of acceleration transducer acquisition are divided into X-axis, Y-axis
And three directions of Z axis, a is used respectivelyx(i)、ay(i)、az(i) come indicate X-axis, i-th of sample on three directions of Y-axis and Z axis
Value, then calculates total acceleration a (i) with formula (1), and formula (2) is recycled to carry out the calculating of acceleration change V (i) later；
V (i)=a (i)-a (i-1) (2)
By way of threshold value is arranged, to determine whether there is the generation that body moves state；If acceleration value V (i) ＞ measured
ε, be considered as current acceleration variation be by body move state change it is caused, if the opposite acceleration value V (i) measured≤
Current value variation is then attributed to noise and caused by ε, and ε is the threshold value of setting；By carrying out experimental analysis to 10 volunteers,
ε is set as 0.05；
The type of action of sleeper is divided into macro motion and micro motion, is judged according to the length of action time of origin；Variation
There are apparent differences on the duration for the big limb motion of amplitude and the small limb motion of amplitude of variation；Big-movement continues
Time is average, and all in 1s or more, the duration of little trick is both less than 1s, so if measuring limb motion duration is less than 1s, just
Think that the limb motion is little trick；Conversely, if measuring limb motion duration is more than 1s, just the limb motion is classified as dynamic greatly
Make；
S2 monitoring of respiration
Monitoring of respiration is judged the respiratory rate of sleeper, because in different sleep stages, the respiratory rate of sleeper
There are significant differences, and at rapid-eye-movement sleep stage (REM), E.E.G variation is rapid, and sleeper often will appear eye and quickly beat
Phenomenon is simultaneously had a dream in this stage, so the respiratory rate in this stage sleep person is usually unstable；In either shallow sleep stage
(Light Sleep), eye jumping phenomena stop, and sleeper's respiratory rate tends to be steady；In deep sleep stages (Deep
Sleep), the respiratory rate of sleeper further slows down；So the size of respiratory rate with the sleep stage residing for sleeper not
It is same and different；
Microphone acquires voice data according to the sample frequency of 8KHz, and voice data is one group according to every 400 sample points and is drawn
It is divided into N group data, N indicates the group number of sample point；Because the sample frequency being arranged is 8kHz, 400 sample points
Time interval is 0.05s, then sets sample data in every group to the average value of this group of data, the purpose for the arrangement is that subtracting
The complexity that will be calculated after few；According to the periodic feature of breath signal, the data after a cycle have similitude, institute
For first of sample point e (l), to obtain e (l)=e (l+T), T indicates the sample point number in a cycle；Normal adult
Respiration rate when quiet is within 16 to 20 times, so setting time minimum period Tmin=16000, the maximum cycle time
Tmax=40000 do a limitation to cycle time, and calculation formula (3) is as follows：
Allow T in TminAnd TmaxIncrease in range, finds corresponding T values when functional value minimum, this makes it possible to be calculated
The duration t=T/8000 of a cycle, corresponding respiratory rate are exactly 60/t；Step-length is arranged when formula (3) calculates
It is 400；
The S3 sounds of snoring monitor
Snoring has certain common features with Breathiness, is Breathiness and the sound of snoring respectively；According to low-frequency range and high band
Ratio come judge acquisition sample sound be Breathiness or the sound of snoring；
Judge the presence of the sound of snoring using the prevailing frequency range of current sound sample segment, is indicated containing n sample with f
The voiced frame to be measured of this point, siF i-th of data value after Fourier changes is indicated, then allowing the ratio of low-frequency range and high band
Value rlh is indicated with formula (4)；
When sleeper is snored by eupnea state shift, sound frequency range will be tilted gradually to low frequency, lead to rlh values
Gradually increase；By analyzing the Snore data more than 100 parts of different sleepers, about settled low frequency becomes larger simultaneously with high frequency ratio
And rlh (f)>Just illustrate that active user is in snoring state when 1.5, it is on the contrary then be in eupnea state；Pass through record
The duration of user's snoring provides snoring duration data when providing sleep quality report, for reference；
S4 illumination monitors
Environment illumination intensity can also have an impact the sleep quality of sleeper；
Since mobile phone front light sensor is closely located to range sensor, it is possible to be with range sensor perception mobile phone
It is no to be blocked, used if perceiving mobile phone and being blocked the last light intensity detected as current intensity of illumination into
Row analysis, when monitoring that mobile phone is not blocked, then detects intensity of illumination again, when intensity of illumination is less than 10Lux, sentences
Break as low light environment；When intensity of illumination, which is more than 10Lux, is less than 100Lux, it is judged as medium light environment；When intensity of illumination is big
When 100Lux, it is judged as strong light environment；For Second Problem, it is contemplated that intensity of illumination will not change frequent, so being
System does not need to real-time monitoring of environmental intensity of illumination；In the method, detection time interval is set to 10 minutes；
S5 sleep states judge
Uncertain feature in being predicted for sleep state carries out sleep state judgement with fuzzy logic operation method；
Detailed process is divided into following steps：
S5.1 collection analysis data；One section of continuous sleep physiological data of acquisition takes this then using 30s as one-shot measurement interval
The monitoring data of each 600s before and after one section of continuous sleep period carry out sleep shape as this section of continuous sleep period
The detection data that state judges；
S5.2 sleep stage fuzzy logic systems generate；This system is to be moved according to body and breathe the two to the judgement of sleep stage
Factor, so extraction does riding Quality Analysis in the average value of this section of continuous sleep period internal respiration frequency, then to body
Dynamic form is compared, and is calculated the average time of macro motion and micro motion generation, is thus obtained the defeated of fuzzy logic system
Enter variable：
X1：The average value and stability of respiratory rate are divided into normal, relatively slow and slow；
X2：Body moves big-movement frequency, is divided into normal, less and few；
The output variable of system is exactly the three phases slept, i.e. F1：Rapid-eye-movement sleep stage (REM), F2：Either shallow sleep rank
Section (Light Sleep), F3：Deep sleep stages (Deep Sleep)；
Thus generating a tool, there are two input variable X1And X2With three output F1、F2、F3Related sleep stage it is fuzzy
Flogic system；
S5.3 membership functions calculate；According to the input variable and output variable set, each input variable of Rational choice
Then membership function carries out integration operation；Triangleshape grade of membership function is selected to carry out analytical Calculation, analytic expression is formula (5)；
S5.4 generates fuzzy logic ordination；The characteristics of being changed using different physiological signals in sleep procedure forms fuzzy logic rule
Then；
S5.5 anti fuzzy methods；Anti fuzzy method is carried out to the result of calculation that previous action obtains, can just obtain wanting result；Not
In the case of having different membership functions with input variable, choosing, there is the result of calculation of maximum membership degree, which to be used as, most terminates
Fruit；
Here it is carry out the overall of sleep judgement based on fuzzy logic theory to realize algorithm；
S6 sleep qualities judge
Sleeper is in sleep procedure the whole night, and the quality of sleep quality is not only related with total sleep time, also by each sleep
The influence of stage distribution proportion, time scale of the sleep quality judgment method dependent on each stage of sleeping, and combine user
People's information and ambient lighting information provide the sleep quality report of individual subscriber after comprehensive analysis.
2. a kind of sleep quality monitoring method based on smart mobile phone according to claim 1, it is characterised in that：It is fuzzy to patrol
Collecting rule specific descriptions rule is：
S5.4.1. in sleep procedure respiratory rate change mechanism；
In REM or awake stage, respiratory rate is very fast and unstable, most slow and flat in the breathing of deep sleep stages
Surely, it is then intervened therebetween in either shallow sleep stage, respiratory rate is relatively slow and relatively steady；
So the changing rule in each sleep stage respiratory rate is:It is awake>REM>Either shallow is slept>Deep sleep；
S5.4.2. in sleep procedure limb motion change mechanism；
The dynamic number of body is relatively more when awake and concentrates on big-movement, and the REM stages are easier to occur to turn over big-movement, and either shallow is slept
The big-movement of dormancy stage is reduced, and is concentrated on this little trick of body twitch, has been arrived deep sleep stages, and limb action disappears substantially,
Even if it is also slight little trick to have；
The changing rule that can obtain moving frequency according to body in sleep and size is ranked up according to above-mentioned judgement is：
It is awake>REM>Either shallow is slept>Deep sleep；
After obtaining the above rule, operation is carried out by the membership function to different input variables, then does final result
Output；If it is few that body moves number, and is little trick, and respiratory rate is slow, then person in servitude of this period for deep sleep
Category degree is just 1.
3. a kind of sleep quality monitoring method based on smart mobile phone according to claim 1, it is characterised in that：With year
Age increases, and the sleeping time of people can continue to reduce, the standard sleep time that each age bracket has oneself special；Carry out sleep matter
Amount judges to first determine whether sleep duration reaches respective standard；Sleep quality score is represented with SQ, full marks are 100 points, if
Reach corresponding sleep duration standard, does not then deduct points, certain score is otherwise subtracted on the basis of total score, 5 points are detained per poor half an hour；
Intensity of illumination can also have an impact sleep quality, after the completion of each sleep procedure, when system monitoring to user is being slept
It when intensity of illumination is higher during dormancy, can also be calculated accordingly, according to 2 points of button, strong light button 5 divides preliminary intermediate light of arranging；Most
After summarize deduction of points item after, it is S to enable the gross score deducted；
Determine that another key factor of sleep quality is the Annual distribution of different sleep stages, the sleep procedure of a high quality
Should be that each sleep stage ratio is reasonable, and is not easy to be disturbed in sleep；Human body can be into deep sleep stages
The recovery of row body function, and then contribute to the formation of the solidification and spatial memory of memory in the rapid-eye-movement sleep phase, therefore want
Comprehensive consideration difference sleep stage just can guarantee the reasonability and validity that sleep quality judges；After considering each factor, most
Sleep quality calculation formula is set as (6) eventually；
Sleep quality marking effect, exactly allow user find in time oneself sleep there are the problem of, from the sleep quality of oneself
In variation tendency, finds the factor for influencing oneself sleep, problem is investigated in time, so as to improve sleep quality.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810455152.9A CN108523901A (en) | 2018-05-14 | 2018-05-14 | A kind of sleep quality monitoring method based on smart mobile phone |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810455152.9A CN108523901A (en) | 2018-05-14 | 2018-05-14 | A kind of sleep quality monitoring method based on smart mobile phone |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108523901A true CN108523901A (en) | 2018-09-14 |
Family
ID=63477364
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810455152.9A Pending CN108523901A (en) | 2018-05-14 | 2018-05-14 | A kind of sleep quality monitoring method based on smart mobile phone |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108523901A (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109199336A (en) * | 2018-09-30 | 2019-01-15 | 深圳个人数据管理服务有限公司 | A kind of sleep quality quantization method, device and equipment based on machine learning |
| CN111077785A (en) * | 2019-11-05 | 2020-04-28 | 珠海格力电器股份有限公司 | Awakening method, awakening device, terminal and storage medium |
| CN112971764A (en) * | 2021-02-20 | 2021-06-18 | 山东大学 | Method and system for detecting waking state and sleeping state of respiratory disease patient |
| CN113724737A (en) * | 2021-08-30 | 2021-11-30 | 康键信息技术（深圳）有限公司 | Method and device for monitoring sleep state, electronic equipment and storage medium |
| CN114145717A (en) * | 2021-12-08 | 2022-03-08 | 四川北易信息技术有限公司 | Sleep state analysis method based on PPG heart rate characteristic parameters and motion quantity |
| CN114176525A (en) * | 2021-12-28 | 2022-03-15 | 深圳市伟晴大健康科技有限公司 | Sleep quality analysis method and device, computer equipment and storage medium |
| CN114732391A (en) * | 2022-06-13 | 2022-07-12 | 亿慧云智能科技（深圳）股份有限公司 | Microwave radar-based heart rate monitoring method, device and system in sleep state |
| CN115120197A (en) * | 2022-06-17 | 2022-09-30 | 歌尔股份有限公司 | Method and device for monitoring breathing condition during sleep, electronic equipment and storage medium |
| CN117531090A (en) * | 2024-01-10 | 2024-02-09 | 深圳市光速时代科技有限公司 | Processing method and system for relieving sleep aiding based on intelligent watch |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103462597A (en) * | 2013-09-18 | 2013-12-25 | 温州医科大学 | Method and device for preventing obstructive sleep sudden death based on intelligent mobile phone control |
| CN104834946A (en) * | 2015-04-09 | 2015-08-12 | 清华大学 | Method and system for non-contact sleep monitoring |
| CN107411704A (en) * | 2017-03-13 | 2017-12-01 | 嘉兴礼海电气科技有限公司 | A kind of sleep quality automatic judging method judged based on Pittsburgh |
-
2018
- 2018-05-14 CN CN201810455152.9A patent/CN108523901A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103462597A (en) * | 2013-09-18 | 2013-12-25 | 温州医科大学 | Method and device for preventing obstructive sleep sudden death based on intelligent mobile phone control |
| CN104834946A (en) * | 2015-04-09 | 2015-08-12 | 清华大学 | Method and system for non-contact sleep monitoring |
| CN107411704A (en) * | 2017-03-13 | 2017-12-01 | 嘉兴礼海电气科技有限公司 | A kind of sleep quality automatic judging method judged based on Pittsburgh |
Non-Patent Citations (1)
| Title |
|---|
| 韩亦俍: "睡眠质量监测系统的设计", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109199336A (en) * | 2018-09-30 | 2019-01-15 | 深圳个人数据管理服务有限公司 | A kind of sleep quality quantization method, device and equipment based on machine learning |
| CN111077785A (en) * | 2019-11-05 | 2020-04-28 | 珠海格力电器股份有限公司 | Awakening method, awakening device, terminal and storage medium |
| CN112971764A (en) * | 2021-02-20 | 2021-06-18 | 山东大学 | Method and system for detecting waking state and sleeping state of respiratory disease patient |
| CN113724737A (en) * | 2021-08-30 | 2021-11-30 | 康键信息技术（深圳）有限公司 | Method and device for monitoring sleep state, electronic equipment and storage medium |
| CN114145717A (en) * | 2021-12-08 | 2022-03-08 | 四川北易信息技术有限公司 | Sleep state analysis method based on PPG heart rate characteristic parameters and motion quantity |
| CN114176525B (en) * | 2021-12-28 | 2023-11-24 | 深圳市伟晴大健康科技有限公司 | Sleep quality analysis method, apparatus, computer device and storage medium |
| CN114176525A (en) * | 2021-12-28 | 2022-03-15 | 深圳市伟晴大健康科技有限公司 | Sleep quality analysis method and device, computer equipment and storage medium |
| CN114732391A (en) * | 2022-06-13 | 2022-07-12 | 亿慧云智能科技（深圳）股份有限公司 | Microwave radar-based heart rate monitoring method, device and system in sleep state |
| US11793415B1 (en) | 2022-06-13 | 2023-10-24 | Yihuiyun Intelligent Technology (Shenzhen) Co., Ltd. | Method, apparatus and system for monitoring heart rate in sleep state based on microwave radar |
| CN114732391B (en) * | 2022-06-13 | 2022-08-23 | 亿慧云智能科技（深圳）股份有限公司 | Microwave radar-based heart rate monitoring method, device and system in sleep state |
| CN115120197A (en) * | 2022-06-17 | 2022-09-30 | 歌尔股份有限公司 | Method and device for monitoring breathing condition during sleep, electronic equipment and storage medium |
| CN117531090A (en) * | 2024-01-10 | 2024-02-09 | 深圳市光速时代科技有限公司 | Processing method and system for relieving sleep aiding based on intelligent watch |
| CN117531090B (en) * | 2024-01-10 | 2024-04-09 | 深圳市光速时代科技有限公司 | Processing method and system for relieving sleep aiding based on intelligent watch |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108523901A (en) | A kind of sleep quality monitoring method based on smart mobile phone | |
| CN107106085A (en) | Apparatus and method for sleep monitor | |
| Watanabe et al. | Noncontact method for sleep stage estimation | |
| Gu et al. | Sleep hunter: Towards fine grained sleep stage tracking with smartphones | |
| Chen et al. | Unobtrusive sleep monitoring using smartphones | |
| JP5961235B2 (en) | Sleep / wake state evaluation method and system | |
| KR102090968B1 (en) | System and method for determining sleep and sleep stages of a person | |
| CN104720748B (en) | A kind of sleep stage determines method and system | |
| JP5740006B2 (en) | Respiration measurement system and REM sleep determination system | |
| CN106846735A (en) | A kind of intelligent mattress warning system | |
| CN109464130A (en) | Sleep householder method, system and readable storage medium storing program for executing | |
| CN108671359A (en) | Sleep householder method | |
| CN107174239A (en) | A kind of sleep monitor | |
| JP2016016144A (en) | Biological information processing system and method of controlling biological information processing system | |
| CN106344034A (en) | Sleep quality evaluation system and method | |
| CN106175699A (en) | Based on magnetic intelligence assisting sleep equipment | |
| CN115831372B (en) | Method, system and device for quantifying and intervening sleep efficiency | |
| CN116211256B (en) | Non-contact sleep breathing signal acquisition method and device | |
| CN114145717A (en) | Sleep state analysis method based on PPG heart rate characteristic parameters and motion quantity | |
| CN105854147A (en) | Intelligent home environment regulation system and method | |
| CN116115198A (en) | Low-power consumption snore automatic recording method and device based on physiological sign | |
| CN106377250B (en) | The tagging equipment of EEG signals data type under sleep state | |
| CN106175698B (en) | Sleep cycle detection device in sleep state analysis | |
| JP2012090913A (en) | Sleep measuring device | |
| Verma et al. | Levels of activity identification & sleep duration detection with a wrist-worn accelerometer-based device |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180914 |
|
| RJ01 | Rejection of invention patent application after publication |